articles

API: /v1/records?type=article

What does Chapter VII of Esenwein and Leeds (1919) say a photoplay title is for, what qualities should it have, and what forms and length do the authors show? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
The authors (1919 revised edition) treat the title mainly as advertising to the public and, with the synopsis, as the editor's first introduction. They adopt Barrett's five attributes (apt, specific, attractive, new, short), say a title may suggest the theme but must not reveal the climax, and describe several title forms: feature-idea titles (girl and deed, locale, two or three persons), fact-stating titles, general theme titles, symbolical double-meaning titles, and (rarely) names. The chapter gives no word or length limit: the only length word is 'short' inside the Barrett quotation; its example titles run 1-5 words. Nothing in the chapter addresses title-card lettering or design (T17); typed-script form for titles is in Ch. VI (cross-reference).
How does Chapter VII say studios treated a writer's title, and which statements are the authors' claims, hypothetical examples or reported cases? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
The authors say writers have complained when editors changed the titles of scripts already accepted and paid for, that editors and directors used to take liberties but that now a studio's change of title 'usually' has a good reason, and they illustrate with a hypothetical (a company featuring only women stars adds a male star, re-works the story and re-titles it, without telling the author). They advise writers to learn each company's lead policy and not to give a script a title that points to a lead of the wrong sex for that company, while keeping the story acceptable to any editor. They also claim companies re-use unoriginal titles because of authors' poor ones. The one actual retitle instance found (in a cross-reference to Ch. V, not a book-wide search) is a footnote on the authors' own specimen. The chapter says nothing about contract, payment for titles or credit.
What does Phillips (1914) say a photoplay title is for, what tests does he give it, and what does he say about its length or form, compared with Esenwein and Leeds on titles? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Phillips treats the title 'chiefly as a commercial asset' and gives it three tests posed as questions: will it attract the hesitating public and the poster in front of the theatre, will it compete among the many releases an exhibitor chooses from, and, 'in a lesser degree', will it appeal to the photoplay editor. He says all three can be met in nine cases out of ten with careful thought, and gives no length, no quality list and no form for titles within lines 846-1025. Esenwein and Leeds (via C3-B286-S94-01) share the advertising-to-the-public function and the editor as reader, add a five-attribute list and a do-not-reveal-the-climax rule, and likewise give no length number. Label: 1914 practice. Excerpts drop Gutenberg italic underscores and collapse line wraps.
What does Phillips (1914) say the cast of characters must contain and who uses it, what format does he show, and how does it compare with the cast list in Esenwein and Leeds? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Phillips says the cast should give each character's individual characteristics where possible and show relationships at a glance, because the director uses it to count players and choose actors by talent and personal type; singles are listed singly and groups or mobs en masse, so the director can judge the outlay in players, costumes and character props. Example 5 shows the format: NAME, a dot leader, a role type such as (Lead) or (Ingenue-lead), another dot leader and a trait phrase. His specimen cast uses a different layout (dash, no dot leaders, role type on only 2 of 6 named entries) and lumps ten minor roles on one line with no scene numbers. Esenwein and Leeds (as relayed) tag minor characters with scene numbers and keep the cast right after the synopsis. Label: 1914 practice. Excerpts drop Gutenberg italic underscores and collapse line wraps.
PROPOSAL: what check could an agent run to propose and test a working title for a 1919-style photoplay script, using only what Chapter VII of Esenwein and Leeds says? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Proposal only (this worker's synthesis of the chapter; nothing was run). 1919 practice from the chapter: a title can come first (a phrase from the Bible, Shakespeare, poets, newspapers or advertisements that supplies the plot-germ), or last, after the synopsis and scenario, by choosing a feature idea; it is judged against Barrett's five attributes, the give-away rule, a worn-phrase list, a released-title lookup and the target company's lead policy, and a quotation title may be repeated as a leader near the climax. The chapter gives no scoring or test; the checklist below is a proposal built from its rules, and the AI use is a hypothesis with no documented use.
What can Demucs (HTDemucs) separate, what can it not, and what are the practical artefact and licence considerations if I try to build an M&E or dialogue stem from a mixed track? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Corrected 2026-10-23 - see C3-B709-01; the original statements below are kept for history and are superseded where the correction says so. Problem: a dub or remix needs a music-and-effects (or dialogue) stem but only a final mix exists. Conditions: single mixed audio file. Method: the Demucs README describes a MUSIC source separation model that outputs four stereo 44.1 kHz stems: drums, bass, other, vocals (a --two-stems option gives e.g. vocals versus accompaniment); an experimental 6-source model adds guitar and piano. v4 is Hybrid Transformer Demucs, trained on MUSDB HQ plus an extra 800-song set, reporting 9.00 dB SDR on the MUSDB HQ test set. What it does not offer (inference from the stem list and training data): no dialogue / effects / ambience classes, so 'vocals' is a singing-voice target, not a speech-dialogue target, and film effects land in 'other' or leak into any stem; the README itself reports 'a lot of bleeding and artifacts' for the experimental piano source. Artefa...
What viewing-comfort practices does Apple's HIG give for video in visionOS? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Immersive video can be uncomfortable when forced on viewers. Conditions. visionOS apps that play video they may not control. Method (HIG). Let people choose when playback starts; use a small playback window that people can resize; keep their surroundings visible during playback; in fully immersive experiences do not let virtual content obscure playback or transport controls; do not start a fully immersive video automatically; provide a thumbnail track (160 px wide thumbnails) for scrubbing; keep inline video 2D and do not expand it to fill a window. Failure modes. Auto-starting immersive playback; occluding controls. Limitations. These are app-level comfort practices, not disparity or convergence limits. Version/date. HIG documentation JSON fetched 2026-10-23.
What are ARKit face tracking's device, range and mesh limits that affect capture for animation or AI characters? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Phone face tracking is a common capture source but has conditions. Conditions. ARKit face tracking on iPhone/iPad. Method. (1) Faces are detected within 3 metres of the front camera. (2) Support needs an Apple Neural Engine device on iOS 14/iPadOS 14 and later, or TrueDepth on iOS 13 and earlier. (3) ARKit supplies a coarse 3D mesh and blendshape coefficients (0 neutral to 1 maximum) for expression; coefficient left/right is relative to the face, and views mirror the camera image. (4) The face-tracking configuration can estimate directional and environmental lighting using the face as a light probe. (5) Apps must include a privacy policy for face data. Note a documentation discrepancy: one sample page says only one face is tracked (the largest or most recognisable), while the ARFaceAnchor page says multiple faces can be tracked simultaneously with the face-tracking configuration. Failure modes (inference from the stated limits). Subjects beyond 3 m, multiple faces in frame,...
How do I use YouTube's 'When your viewers are on YouTube' report and the other Audience-tab reports to choose publish and live-stream times and topics? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: choosing when to publish or go live without folk wisdom. Conditions: YouTube Studio > Analytics > Audience tab; data windows are fixed by the report. Method (documented): 'When your viewers are on YouTube' shows when your viewers are online across YouTube in the last 28 days and is suggested for building community, picking a Premiere time and planning a live stream. Related Audience-tab reports: monthly audience (unique viewers in the previous 28 days, recalculated daily); channels your viewers watch outside yours (28 days) for collaboration ideas; other videos, Shorts, live streams and podcasts they watched outside your channel (past seven days) for topics, titles and thumbnail ideas; and a formats-watched report (28 days) for deciding between videos, Shorts and live. Failure modes: treating the online-time grid as proof that publishing at that time raises views; the page makes no such claim. Limitations: the page reports no evidence that upload timing changes performance;...
How do Kling motion control (reference video) and the brush-based motion paths work: inputs, orientation modes, durations, limits and failure modes? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Directing a character's body and face movement with a reference clip instead of text. Conditions. Kling 2.6/3.0 motion control as exposed by fal, Replicate and Kling's own guide (hosted, proprietary). Method. (1) Reference-video motion control (fal 2.6 endpoint): image_url (character image: clear body proportions, avoid occlusion, occupy over 5% of the image), video_url (realistic character with visible upper body or full body including head), character_orientation = 'image' (keeps the image's pose and facing; max 10 s; good for camera-movement goals) or 'video' (transfers motion and orientation literally; max 30 s; for dance and choreography), optional prompt, keep_original_sound. (2) The prompt is scene-setting, not motion: describe identity, environment and style; over-describing motion is the most frequent error. (3) Kling's guide: motion can come from an uploaded video or a motion library; VIDEO 3.0 adds element binding to improve facial consistency across angles and o...
What rules does Don Carson's environmental-storytelling essay give for a consistent, readable story world? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Spaces that look good but tell nothing, or break their own rules. Conditions. Game or themed-environment design (essay from March 1, 2000, drawing on theme-park practice). Method. (1) Define a big-picture story as a set of rules guiding design and team; never break those rules. (2) Answer 'Where am I?' within the first 15 seconds, then 'What is my relationship to this place?' through clues rather than text dumps. (3) Use cause-and-effect vignettes: staged areas that let the player conclude what happened (broken doors, a crashed vehicle, charred remains), also to show time passing or player-caused changes. (4) Follow 'bread crumb' trails left by a fictional predecessor. (5) Give familiar anchors in alien environments. Failure modes. Adding elements just because they are cool; alienating all-alien levels. Limitations. Craft essay with examples from 1990s games and attractions, not empirical evidence. Version/date. Page dated March 1, 2000; fetched 2026-10-23.
What synchronisation and content-authoring constraints does Epic give for ICVFX (genlock, video playback, frame-rate target)? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Multiple render nodes drive one wall and must show identical frames. Conditions. nDisplay cluster with camera, tracking and render nodes. Method. (1) Each device (camera, computers, tracking) has its own internal clock; without unification you get problems such as tearing, which genlocking with nDisplay prevents; engine timecode and frame generation must also match the camera. (2) Do not use video files as animated surfaces: video encoding gives no method to keep every node on the same frame; use texture flipbooks or an EXR sequence driven by a level sequence played simultaneously. (3) Performance: Epic's test content targeted 48-72 fps on artist workstations at 4K full screen as a benchmark, then profiled on machines matching the heaviest on-stage render node with a proper inner and outer frustum. (4) Avoid very large textures and dense meshes where not needed. Failure modes. Content that plays fine on one workstation but desyncs on the cluster. Limitations. The 48-72 fps ...
On an LED volume, what are the inner and outer frustums, which one moves with the camera, and what is the green-screen fallback? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. LED-volume shots mix a camera-tracked image with ambient lighting content. Conditions. In-camera VFX with Unreal nDisplay (Epic docs). Method. (1) The inner frustum is the part of the wall showing what the virtual equivalent of the physical camera should see, set by current lens focal length and tracked as the camera moves, which gives parallax from a full 3D world rather than a flat plate. (2) The outer frustum is everything displayed outside the camera's field of view; it turns the panels into a light and reflection source for the physical set and stays static when the camera moves, as real-world light does. (3) Each shot setup can be placed at a chosen location in the virtual environment to decide what the outer frustum emits. (4) For shots that cannot be in-camera finals, the inner frustum can be changed to green screen with tracking markers while the outer frustum keeps lighting the set, which also limits green spill. Failure modes. Reflections and light that do not ma...
How should detail, light and props be placed to direct attention in a story space, per Carson? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Over-detailed, evenly lit spaces hide story cues. Conditions. 3D level or set design. Method. (1) Use contrast: a large room with one shaft of light on a solitary prop beats a room full of detailed elements; make other props supporting players to the important story element. (2) Use fewer, quieter textures and save the most decorative element for the end of a corridor to pull the viewer on. (3) Avoid flooding a level with light, which flattens depth; let large areas fall into shadow while not hiding vital game elements. (4) Do not offer too many choices at once. (5) Treat texture maps as stage canvas, not wallpaper. Failure modes. Lighting that reveals every nook; cluttered high-frequency textures. Limitations. Craft guidance from an era of limited rendering; principles are compositional. Version/date. Essay dated March 1, 2000; fetched 2026-10-23.
How should I phrase motion in text prompts for Kling-style models (visible cues) and what does Hailuo's 'one major motion per clip' guidance claim about ghosting? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Text-only motion prompts give vague or ghosted movement. Conditions. I2V/T2V prompts of 4-15 s clips. Vendor statements. Kling's guide: motion prompts work best when they describe what the viewer can see: posture, speed, direction, ground contact, camera movement, clothing motion and environmental response; replace the single verb 'running' with speed, effort, posture, surface contact and camera relationship; Kling 3.0 supports up to 15 s with multi-shot. Hailuo's page (creator-style claims): keep one major motion per clip - if the subject moves fast keep the camera static or use a simple slow push-in, because combining a sprint with a dolly zoom or whip pan breaks temporal logic; it also describes ghosting as trailing limbs. Evidence quality. Hailuo labels its 'Velocity-to-Artifact' model as hypothetical and not lab-tested, built from internal pattern recognition. Method (inference). Write each motion as 3-4 visible cues; assign only one dominant motion (subject or camera)...
What tasks does Wan VACE cover (reference-to-video, video editing, masked editing, expand), what inputs does it take, and which base-model licences apply? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. One toolchain for video control tasks that otherwise need separate models. Conditions. VACE built on Wan2.1 (1.3B at about 81x480x832, 14B at about 81x720x1280) and LTX-Video 0.9. Method. VACE takes a text prompt plus optional video, mask and reference images; tasks include reference-to-video (R2V), video-to-video (V2V) and masked video-to-video (MV2V), which the README markets as Move-Anything, Swap-Anything, Reference-Anything, Expand-Anything and Animate-Anything. Run through vace_pipeline.py with a preprocessor task name, e.g. --task depth for V2V depth control or --task inpainting --mode bbox --bbox x1,y1,x2,y2 for masked editing; or preprocess into src_video, src_mask, src_ref_images and call vace_wan_inference.py. Prompt extension is needed for LTX-Video and English-language Wan2.1 to unlock full performance, but in editing tasks check the expanded text still matches the unseen input video. Failure modes. The README warns prompt extension may not match the video bein...
What are the camera-tracking options for virtual production and what does the virtual camera need to match? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. The wall image is only correct if the virtual camera matches the physical one. Conditions. In-camera VFX shoots. Method. (1) Tracking relays the physical camera position and movement to the engine so the correct perspective is rendered. (2) Common methods: optical (IR cameras tracking reflective or active markers), feature tracking (image patterns instead of markers), and inertial (IMU gyroscope and accelerometer, often combined with optical or feature tracking). (3) Epic recommends using multiple sources, such as optical plus inertial, to improve tracking data. (4) Live Link distributes tracked camera data, and with nDisplay the primary node replicates it to the cluster. (5) Calibration: the virtual camera's position and orientation must closely match the physical camera, and a Camera Calibration plugin generates data to align the cameras and model lens distortion for post-processing. Failure modes. Lens distortion or nodal offset not calibrated; unsynchronised tracking la...
How do I avoid moire when filming an LED wall, and which variables does Epic say control it? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Visible interference patterns ruin LED-volume shots. Conditions. Camera pointed at an LED wall. Method. (1) Understand the cause: moire appears when display pixels and camera sensor pixels are slightly offset. (2) Distance from the screen, LED pixel pitch and camera sensor size together determine how far the subject must be from the wall to avoid visible artifacts. (3) Keep the camera focus plane in front of or behind the LED surface so the wall is slightly out of focus; moire can appear when the focus plane aligns with the panels. (4) Keep the camera perpendicular to the wall where possible, since severe angles make moire more likely. Failure modes. Focusing on the wall (for example a rack focus onto the background); steep angle shots. Limitations. Epic gives no numbers (no minimum distance or pixel-pitch table); those must come from stage test shots. Version/date. UE docs fetched 2026-10-23.
What does Apple say about spatial metadata and disparity adjustment when converting side-by-side 3D to spatial video? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Stereo files can be uncomfortable or treated as plain stereo. Conditions. Side-by-side stereo video converted to MV-HEVC for Apple Vision Pro. Method. (1) Convert by splitting each frame and writing left and right halves to the left- and right-eye layers. (2) Spatial metadata describes the left/right cameras (baseline, horizontal field of view, disparity adjustment). Adding it makes Apple platforms treat the file as spatial rather than just stereo and enables visual treatments that reduce common stereo discomfort. (3) Sample values: virtual cameras 64mm apart with 80 degree horizontal field of view; a disparity adjustment of +2% of image width gave a comfortable result in a window, placing the nearest object just behind the front of the window while keeping depth. Failure modes. Wrong baseline or field of view metadata for the content; nearest objects in front of the window plane. Limitations. The values belong to Apple's sample content; other content needs its own choices....
What do YouTube's playlist and Posts (community post) help pages document that matters for series packaging and audience contact, and what do they not say about session watch time? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: planning a series playlist and companion posts. Conditions: desktop YouTube and Studio. Method (documented): playlists can hold videos and Shorts; creators can reorder them by dragging (not for Shorts-only playlists), filter by All/Shorts/Videos, and there is a daily limit on public playlists a channel can create, raised by advanced features. Deleting a playlist removes its URL and title from Analytics and stops attributing watch time to it, though the watch time stays in broader reports. Posts can be text, image/GIF (up to 10 images, JPG/PNG/GIF/WEBP, up to 16 MB, a suggested 1:1 ratio because that is how the feed shows them), video, poll, quiz or playlist posts; they can be scheduled; mentions with @ may notify the channel; a post sharing another creator's video may notify them; a posts-per-24-hours limit applies; and a post about a video already in a viewer's feed might not show again. Failure modes: expecting analytics for a playlist after deleting it; sharing a viewer'...
What may other creators reuse from my Shorts (audio, Cut, Green screen, Collab, templates), how long can a sampled segment be, and how do I opt out? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: knowing what rights your Shorts carry once published, and what happens to derivatives when you restrict or delete. Conditions: YouTube Shorts remix tools on mobile; Studio for opt-out; Content Manager partners have separate controls. Method (documented): other creators can remix audio from Shorts and long-form videos ('Use this sound'), make a Cut, use a video as a Green screen background, Collab, or use a Short as a template; remixes are credited back to the source. Music-video limits: Cut samples a 1-5 second segment (audio and video); Sound samples up to 90 seconds of audio, and some partner agreements limit this to 30 seconds. Default: Shorts are automatically opted in to remixing; a creator can opt Shorts out of video remixing in Studio (Content > video > Show more > Shorts remixing), also in bulk and at upload; only Content Manager partners can opt out of both audio and video. Effects of restricting or deleting: existing audio remixes are muted, set to Unlisted and sc...
What four ways does Henry Jenkins say game spaces carry story (evocative, enacted, embedded, emergent), and how do they combine? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. The ludology/narratology dispute stalls practical design. Conditions. Designing spatial or open-world narrative. Method. Jenkins proposes game designers as narrative architects. Environmental storytelling sets the preconditions for narrative in at least one of four ways: spaces evoke pre-existing narrative associations (genre/story worlds), act as a staging ground for enacted events, embed narrative information in the mise-en-scene for players to discover, or supply resources for emergent narratives (his example: The Sims). Detective-style design: a loosely structured story driven by the player's exploration plus a pre-structured one embedded in the space awaiting discovery; mixing enacted and embedded elements balances interactivity flexibility with coherence of a pre-authored narrative. Failure modes. Treating a space as empty of narrative because it lacks cutscenes. Limitations. Essay from the early 2000s with game examples (Myst, Half-Life, The Sims); fetched text carri...
Does a rewritten post title change engagement when the same YouTube video is shared on Reddit, and what text features go with higher engagement? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: separating title wording from video popularity when judging packaging. Conditions: Hu et al. (Georgia Tech, SRI; arXiv 2505.03769, ACM manuscript) built a dataset of 24.1 million Reddit posts linking YouTube videos (19.8M video posts, 12.6M unique videos, 305K subreddits). They report that 21% of post titles are minimally modified from the video title. Method: controlled post pairs sharing the same or similar videos within a 30-minute window, plus an 'inverse' set where the less-popular video's post got more engagement; the mixed controlled set has 86,035 pairs. Measured result: title rewrites measurably relate to engagement; more engaging rewrites tend to show emotional resonance, lexical richness and alignment with community norms; a fine-tuned BERT classifier ranked titles at 74% accuracy in the controlled setting, while GPT-4o was only slightly better than chance. Failure modes: this measures Reddit votes and comments on titles, not YouTube clicks or watch time. Limitat...
How should an AI world or scene generator place story cues, combining Carson's and Jenkins's craft? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Generated scenes are often visually dense but narratively empty or contradictory. Synthesis (inference, built from findings 11 to 13, not from a source recommendation). (1) Start from a rules statement for the world and check every generated element against it (Carson: do not break the rules). (2) Make the first view answer where the player is within about 15 seconds, using one dominant landmark and a few genre-evoking cues (Jenkins: evocative spaces). (3) Place a small number of cause-and-effect vignettes with clear traces of a prior event, each with a single highlighted focal prop (Carson: vignettes, single shaft of light). (4) Tag each cue as evocative, enacted, embedded or emergent and make sure embedded cues are consistent with the story ledger and discoverable. (5) Reserve decoration to lead the eye toward the next space; avoid evenly lit clutter. (6) Keep a list of cues so later scenes can alter them to show time passing. Failure modes. Random props that contradict t...
For FLUX models, which variants use true CFG and negative prompts, and which are guidance-distilled with a scalar guidance_scale? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Negative prompts written for SD models are carried over to FLUX. Conditions. diffusers FluxPipeline family. Source statements. FLUX.1-schnell is timestep-distilled: guidance_scale needs to be 0, it benefits from fewer steps and max_sequence_length cannot exceed 256. FLUX.1-dev is guidance-distilled: it takes about 50 steps and the docs use guidance_scale=3.5. Fill and control variants (Canny, Depth, Fill) are also guidance-distilled; the docs use guidance_scale=30.0 for Canny, 10.0 for Depth. One diffusers example for FLUX.1-dev with an IP-Adapter passes negative_prompt='' together with true_cfg_scale=4.0, showing that FluxPipeline exposes a separate true CFG path. BFL's CLI example for Kontext uses --guidance 2.5 (C3-B576-16). Method (inference). Treat the scalar guidance_scale on dev as a distilled 'fake CFG' that is not driven by a negative prompt; to use a negative prompt on FLUX.1-dev set true_cfg_scale above 1 and supply negative_prompt, accepting about double compute...
How do I run Wan2.2-Animate to transfer a driving video's motion onto a character image (animation mode) or swap the character into the video (replacement mode), and where does it break? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Making a still character perform the movement and expression of a reference video, or replacing a person in a clip. Conditions. Wan2.2-Animate-14B (weights Apache 2.0; diffusers port 2025-11-13). Inputs: one character image plus one driving video. Method. (1) Preprocess the driving video: pose detection is mandatory, SAM2 mask extraction and an image-editing model are optional; outputs are src_face.mp4 and src_pose.mp4 (animation mode) plus src_bg.mp4 and src_mask.mp4 (replacement mode). Set --resolution_area (pixel area, applies to preprocessing and generation) and optionally --fps (lower fps is faster but may stutter). (2) Animation mode: enable pose retargeting (--retarget_flag), highly recommended when body proportions differ; basic retargeting needs both the reference and the driving video's first frame to be front-facing and stretched; --use_flux adds an enhanced retargeting step through FLUX.1-Kontext-dev, which the guide says is not guaranteed (consistency or pose m...
How do I run a small, honest listening test with hidden reference and anchors (ITU-R BS.1534 MUSHRA), and what does that method not cover? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: compare several audio versions (codec, denoiser, TTS, separation) with a defensible test. Conditions: audio differences are in the small-to-medium impairment range (BS.1534 is for intermediate quality; BS.1116 is for small impairments). Method (BS.1534-3): present listeners with a known reference, several systems under test, plus hidden reference and anchors in one trial; the standard anchor is the original low-pass filtered at 3.5 kHz and the mid-quality anchor at 7 kHz; keep a trial to no more than 12 signals; use experienced listeners; post-screen assessors by excluding those who give a very high grade to a heavily impaired anchor or frequently grade the hidden reference as impaired (the Recommendation gives numeric thresholds of 15% of items, e.g. reference below 90 or mid anchor above 90); train listeners on the range of impairments first; analyse with the statistics in the annexes (confidence intervals; non-parametric options). The Recommendation also notes the name M...
In spatial/360 audio, what changes when the listener's head moves: which parts rotate with the head, which stay fixed, and how many channels do the orders need? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: planning audio for 360/VR/head-tracked playback (e.g. a need about head movement). Conditions: ambisonic (scene-based) delivery decoded to binaural headphones. Method: ambisonic scenes are decoded and rendered binaurally, and the whole sound field is rotated by the head/camera orientation; Omnitone, Google's Web Audio decoder, applies a 3x3 or 4x4 rotation matrix each animation frame (it can be wired to a three.js camera or device orientation) to rotate the entire sound field. Channel counts by order are 4 (first order), 9 (second) and 16 (third) channels in that library. YouTube's spatial audio spec uses AmbiX (ACN ordering, SN3D normalisation), first order only, W,Y,Z,X at 48 kHz in one audio track, and it also supports a 6-channel FOA plus head-locked stereo (L,R) layout, where the stereo pair does not change when the viewer moves their head. Production consequences: put narration, dialogue-over-picture, music beds and UI sounds in the head-locked pair; place only source...
Which SSML tags does Amazon Polly support for neural, long-form and generative voices (is emphasis available)? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: SSML written for one voice type silently loses effect on another. Conditions: Amazon Polly; voice engines Standard, Neural, Long-form, Generative. Method: Polly's page says all tags except <amazon:domain name="news"> are supported for Standard voices, then tabulates the others by voice type. From the table as fetched: <break>, <lang>, , <s>, , <w>, <speak> are fully available for neural, long-form and generative voices; <emphasis> is not available for any of the three; <prosody> is only partially available for all three (and <prosody amazon:max-duration> is not available); and <phoneme> are partial for generative voices; <amazon:auto-breaths> and <amazon:effect phonation=soft> are not available for the three newer engines; the dynamic-range-compression effect works for neural and long-form but not generative. Use (inference): when moving from Standard to neural/generative voices, replace emphasis with wording/punctuation, use <break> and /<s> for rhythm, a...
What adaptation techniques do dubbing translators use to reach lip-sync versus isochrony, and what do they give up in fidelity? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: a literal translation rarely fits the mouth or the clock. Conditions: any dubbed dialogue line. Method: Chaume's techniques as reported: for lip-sync, repetition (cognates), word-order shift so the marked sound lands on a similar sound, substitution with a synonym/antonym/hypernym, reduction or amplification, and omission or addition. For isochrony, amplification (repetition, gloss, paraphrase, longer synonyms) or reduction (ellipsis of modal verbs, interjections, phatic markers, social formulae, vocatives and proper names; omission of redundancy with the image; deictics and all-purpose words). Fidelity trade-off: such omissions and additions would count as errors in written translation but are permitted in dubbing. Example (measured, 10 segments of a Persian dub of The Green Mile): repetition, word-order shift and substitution dominated lip-sync, while isochrony used reduction 58 percent and amplification 42 percent; amplification was also used to disambiguate. Koveriene f...
How do I measure and fix a constant audio offset in an edit with FFmpeg (-itsoffset, adelay, aresample async), and what does each tool actually change? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: separately generated or dubbed audio lands early or late against picture. Conditions: a CONSTANT offset has been measured (clap/flash marker, or a transient in both streams); drift is a different problem. Method (all statements from the FFmpeg docs unless marked): -itsoffset is an input option; the offset is added to the input's timestamps and a positive value delays that input's streams, so place it before the -i of the audio file to delay audio (a negative value to advance it is my inference from 'added to the timestamps'; test it). The adelay filter delays selected channels and fills the gap with silence; delays are in milliseconds per channel separated by '|', with an S suffix for exact samples and 'all=1' to apply the last delay to all remaining channels, e.g. adelay=delays=64S:all=1. aresample with async can stretch/squeeze samples to follow timestamps (docs example async=1000, maximum 1000 samples per second of compensation), which is a drift-handling tool, not an of...
What do Chaume's three dubbing synchronies (lip, kinesic, isochrony) each require, according to translation-studies sources? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: dubbing adaptation is judged on several kinds of match at once, but briefs often say only 'sync'. Conditions: dubbing (not subtitles or voice-over) of dialogue that is on screen. Method: Chaume's typology, as reported in the open-access sources, has three parts. Lip synchrony adapts the translation to the on-screen character's articulatory movements, above all bilabial and labio-dental consonants and open vowels. Kinesic (kinetic) synchrony keeps the words logically consistent with the characters' gestures and body movements (for example a nod must not accompany 'no'). Isochrony matches the duration of the target speech phrases and pauses to the source; its absence is called the most noticeable sign of a poorly dubbed product, and the Persian study says the viewer notices when lips have closed but speech continues, or lips move in silence. Chaume's 2012 book defines synchronisation as the process of matching the translation to actors' body and articulatory movements in a st...
What exactly goes into ComfyUI's output-cache key for a node, so I can predict whether editing one widget re-runs a node (input signature plus all ancestors, not node IDs)? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Agents that edit an API-format prompt and resubmit need to predict which nodes re-execute. Conditions. ComfyUI v0.42.5 (tag, commit 483dae43), any cache mode except --cache-none; the output cache is built with CacheKeySetInputSignature in comfy_execution/caching.py. Method. For each node, get_node_signature builds a list from the node itself plus every ancestor (get_ordered_ancestry). Each immediate signature is [class_type, IS_CHANGED/fingerprint value] plus, only when the class sets NOT_IDEMPOTENT or declares a UNIQUE_ID hidden input, the node ID; then every input key in sorted order: literal widget values are included verbatim, linked inputs are replaced by ('ANCESTOR', ancestor_order_index, socket). The list goes through to_hashable. Consequences: (a) changing any literal upstream changes the key of every downstream node; (b) node IDs are not part of the key for ordinary nodes, so two identical sub-graphs with different IDs share one cache entry and renumbering nodes do...
How exact must phonemes match for lip-sync in dubbing, and what do open-vowel, bilabial and viseme substitutions allow? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: judging whether a dubbed word 'fits' the mouth. Conditions: visible mouth, usually close-up. Method: the guidance targets visually distinct sounds rather than identical ones: bilabials (open/closed lips), labio-dentals and open vowels are the most visible. Substitutions are allowed within a class: open vowels can be swapped, and a /p/ can be replaced by /m/, /b/, or the labiodentals /f/ or /v/. The viseme theory formalises this by grouping phonemes by lip shape; rounded and protruded vowels form the 'fourth viseme'. Example data (one Lithuanian dub of an English animated film, 476 stressed rounded/protruded vowel cases, 1 h 35 min): the translator used the fourth viseme in 46 percent of cases, the third in 32 percent, the second in 12 percent and the first in 10 percent; stress matched in 68 percent of cases. When a rounded vowel could not be matched, first- and second-viseme sounds fitted better than third-viseme sounds. Failure modes: matching spelling or phoneme identity...
In an open TTS evaluation, what are WER and speaker-similarity (SIM) computed with, and why can a lower WER mean worse speaker similarity? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: judge a TTS or voice-cloning output with objective numbers. Conditions: zero-shot TTS evaluated on a fixed test set. Method: the Seed-TTS evaluation (repo and paper) computes WER by transcribing the generated audio with an ASR (Whisper-large-v3 for English, Paraformer-zh for Mandarin) and comparing with the intended text, and SIM as the cosine similarity between speaker embeddings from a WavLM-large model fine-tuned for speaker verification, comparing each synthesized utterance with its reference clip. The released set uses 1,000 Common Voice samples (English) and 2,000 DiDiSpeech-2 samples (Mandarin); scripts are cal_wer.sh and cal_sim.sh. The paper cautions that a lower WER does not necessarily mean better subjective speaker similarity: lower WER often means more 'standardized', less accented or less expressive speech that the ASR finds easy. Use (inference): report WER and SIM together with a listening test; name the ASR and embedding models and versions, because scores ...
Which cache classes does ComfyUI construct for each --cache-* flag, what is the separate 'objects' cache, and how do I clear cached node results on a running server? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Memory growth or stale-looking reruns on a server; choosing a cache flag. Conditions. ComfyUI v0.42.5. Method. main.py prompt_worker maps flags to execution.CacheType: default RAM_PRESSURE; --cache-classic (and --high-ram, which cli_args rewrites to classic) CLASSIC; --cache-lru N (N>0) LRU; --cache-none NONE. The flags live in one mutually exclusive argparse group. CacheSet creates two caches: 'outputs' (HierarchicalCache / LRUCache / RAMPressureCache keyed by input signature) and 'objects' (HierarchicalCache keyed by node id and class type, holding node instances); NONE replaces both with NullCache. --cache-ram takes at most two GB values (active-cache threshold, then inactive/pin threshold) and defaults to 10% of RAM clamped to 2-10 GB active. Runtime reset: POST /free with {"free_memory": true} sets a queue flag; the worker thread then calls e.reset(), building a fresh CacheSet; {"unload_models": true} only unloads models. Failure modes. /free only sets flags and return...
What do break, emphasis and prosody mean in W3C SSML 1.1, and how much is a TTS engine actually obliged to do with them? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: control pause, stress, rate and pitch of synthetic speech portably. Conditions: engines that accept SSML. Method: <break/> is an empty element; strength takes none, x-weak, weak, medium (default), strong or x-strong, and time takes a duration (e.g. 250ms or 3s); if neither attribute is given the processor produces a break stronger than it otherwise would; if both are given the time sets the duration and the strength shapes other prosodic changes. <emphasis> requests prominence, but the engine decides how to render it because emphasis differs between languages, dialects and voices. <prosody> controls pitch, contour, range, rate, duration and volume; rate may be a percentage multiplier (100% = unchanged), volume a signed dB value (-6 dB roughly halves amplitude) or a label, and it is an error to use prosody with no attributes. The spec says all prosodic attribute values are indicative (the engine may clamp or substitute) and that authors are encouraged not to casually mix the...
How do Azure AI Speech and Google Cloud Text-to-Speech document emphasis, prosody and style control, and what are the caveats for HD or newer voices? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: choose markup that the target voice honours. Conditions: Azure Speech SSML and Google Cloud TTS SSML as documented. Method (Azure): <prosody> adjusts pitch, contour, range, rate and volume (contour takes percent/pitch pairs); <emphasis> word-level stress tuning is documented only for en-US-GuyNeural, en-US-DavisNeural and en-US-JaneNeural, and the page warns low-pitch, short words may not be raised enough to notice. For HD voices (DragonHD family) expressive control moves to styles: <mstts:express-as style=...> or inline style markers such as [whispering]; a marker applies to all later sentences until reset with [Neutral], and \r\n, and automatic sentence boundaries reset to default style while <break> does not. Style degree 0.01-2 is ignored if unsupported for the voice. (Google): the page states that not all elements of the W3C spec are supported; <emphasis> levels are strong, moderate, none, reduced; it advises using <emphasis> and <prosody> only around full sentence...
What lip-sync (sound/picture) timing tolerance does EBU R37 set for a broadcast programme, and what is the stage-by-stage versus end-to-end figure? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: an edit or AI-dubbed/AI-generated-audio deliverable needs a defensible number for 'is the audio in sync'. Conditions: broadcast or broadcast-style delivery; R37 is a 2007 EBU recommendation measured under EBU R28 viewing conditions. Method: R37 gives two figures. (1) Accuracy at each stage of the chain: audio from 5 ms early (sound before picture) to 15 ms late (sound after picture). (2) At any output intended for emission, the overall end-to-end difference should lie within Table 1: sound before picture up to 40 ms, sound after picture up to 60 ms. It says the values come from subjective tests of when lip/speech asynchrony becomes perceptible to 50% of observers, that perceptibility depends on content and viewing distance, and that HD needs particular care. It also says audio and video should be kept coincident in the file and not advanced or retarded to compensate for downstream delays, and that concurrent flash/blip (1 kHz tone, one frame) header markers are the measurem...
What does MMAudio take as input, what duration does it work at, what hardware does it need, and where does it fail (speech, unfamiliar sounds)? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: choose or set up a video-to-audio (foley) generator for a silent AI video clip. Conditions: a short clip, optionally a text prompt. Method (repo README): demo.py takes --video and --prompt (omit --video for text-to-audio) with --duration=8; the default output and training duration is 8 seconds and a large deviation from it may lower quality, so longer clips need segmenting and crossfading (inference). Inference is reported at about 6 GB GPU memory in 16-bit mode; the authors tested only on Ubuntu. Known limitations listed by the authors: sometimes unintelligible speech-like sounds, sometimes background music, struggles with unfamiliar concepts (can generate gunfire but not an RPG). The paper adds that it generates 'unintelligible mumbles' when asked for human speech from mouth movement or text because it targets general audio effects (Foley). Implication (inference): use it for foley/ambience beds under a separate dialogue track, never for lines; audit the output for stray ...
What exact numbers does Netflix set for subtitle in-times, out-times, shot changes, minimum duration and gaps (at 24 fps and other frame rates)? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: timing rules for subtitles in frames, not just characters per second. Conditions: Netflix timed text at any frame rate; the page writes rules for 24 fps and scales the half-second rule (12 frames at 24 fps, 15 at 30 fps, 30 at 60 fps) while the minimum gap stays 2 frames. Method (source statements): in-time on the first frame of audio, within 1-2 frames acceptable; out-time ideally at least half a second after audio ends when no subtitle follows; shot-change rules: dialogue starting on or within half a second past the cut sets the in-time to the first frame of the shot; an out-time within half a second of the last frame before a cut extends to the cut, respecting a 2-frame gap; subtitles may cross a shot change only if the dialogue does; minimum gap between subtitles 2 frames, and gaps of 3-11 frames at 24 fps must be closed to 2 frames ('chaining'), so gaps are either 2 frames or at least half a second. Minimum duration is 20 frames (stated as 4/5 s on this page, which doe...
What do ITU-R BT.1359-1 detectability and acceptability thresholds say about audio leading versus lagging picture, and which sign convention does it use? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: viewers tolerate audio lag and audio lead very differently, so a single symmetric +/- figure misleads. Conditions: broadcast television; thresholds are averages from subjective tests, original (1998) Recommendation. Method: BT.1359-1 states detectability thresholds of about +45 ms to -125 ms and acceptability thresholds of about +90 ms to -185 ms on average, where a POSITIVE value means sound is advanced (early) relative to vision. It recommends an overall sound/picture tolerance of +90 / -185 ms, +25 / -100 ms between image source and the zero reference, and +22.5 / -30 ms from final source selection to the transmitter input, with no downstream segment outside the broadcaster's control adding more than +/-2 ms. Its appendix explains the 22.5/30 ms limits as half a grade point above the detectability threshold on the five-point scale (one grade is about 60 ms for sound delay, about 45 ms for sound lead). Practical reading (inference): audio that lags picture is detected lat...
If I overwrite a checkpoint or LoRA file in place under the same filename, will ComfyUI's cache notice? Which core nodes hash file contents in IS_CHANGED? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. A long-lived headless server may keep serving cached results after a model file is replaced under the same name. Conditions. ComfyUI v0.42.5, default RAM-pressure or any non-null cache, same server process. Method. In nodes.py only three IS_CHANGED definitions exist (LoadLatent, LoadImage, and the channel-based LoadImageMask delegating to LoadImage); the first two hash the file with sha256. CheckpointLoaderSimple has no IS_CHANGED and its only input is the ckpt_name string chosen from folder_paths.get_filename_list. From the cache-key code (previous finding) the node signature therefore contains just the filename. Inference: replacing the file's bytes without renaming does not change the key, so a still-cached loader (and anything downstream) can be reused; use new filenames per version, or restart the server, or POST /free with free_memory true to reset the executor caches (main.py prompt_worker calls e.reset()). Failure modes. Provenance logs that record only the filename...
What focal-length vocabulary and lens-type distinctions (wide, telephoto, prime, zoom, rack focus) do film-course texts use, and how do I place them in a video prompt? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Lens words in prompts need craft-correct meaning. Conditions. Image and video prompts for Veo-class and other models. Course-text statements (LibreTexts, Nashville State CC film course). Focal length sets angle of view and magnification: shorter = wider and smaller magnification, longer = narrower and more magnified. Rule of thumb in the text: below 35 mm is wide-angle and above 70 mm telephoto. Cinematographers almost always use primes; zooms change focal length within the lens. Depth of field is the range of distance in sharp focus; narrow DoF isolates a subject, deep focus keeps everything sharp; changing DoF within a shot is rack/pull focus. Google's Veo 3.1 guide lists lens and focus terms to use in the [Cinematography] slot: shallow depth of field, wide-angle lens, soft focus, macro lens, deep focus. Method (inference). Name lens type plus an effect in one clause ('wide-angle lens, deep focus' or 'telephoto lens, shallow depth of field'); use 'rack focus' only with a ...
How do IS_CHANGED and fingerprint_inputs really work in ComfyUI: what do they receive, when are they evaluated, what happens if they raise, and where do the docs differ from the code? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Custom-node authors and agents rely on IS_CHANGED to force or avoid re-execution. Conditions. ComfyUI v0.42.5; class-level IS_CHANGED (V1 nodes) or fingerprint_inputs (V3 nodes). Method. IsChangedCache.get in execution.py looks first for a V3 'fingerprint_inputs' override, else an IS_CHANGED attribute; if neither exists the value is False. It is called with get_input_data(node inputs, ..., None), i.e. only literal widget values: the code comment says outputs of upstream nodes are intentionally not used, so linked inputs are not available to IS_CHANGED. The result list is stored on the node dict (computed once per prompt) and enters the cache signature. Return a stable content hash (the built-in LoadImage returns a sha256 of the file) to invalidate on content change; return float('nan') to force re-execution every run. Failure modes. If IS_CHANGED raises, the code logs a warning and stores NaN, which silently forces re-execution every time. Returning True/False is a trap bec...
How is a CMOS (comparative MOS) test set up in a recent TTS paper, and what threshold is treated as no difference? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: decide whether two voice outputs differ audibly. Conditions: listeners compare a synthesized clip against a human or other-system clip for the same text. Method (as described by Seed-TTS): for each test sample evaluators first hear a reference clip of the target speaker; then hear the synthesized output and the ground-truth human recording in random order; they rate which has higher speaker similarity and expressiveness on a scale from -2 to +2 (the extremes mean least and strongest preference for the first sample); results are rearranged into one direction (system versus human) and averaged over evaluators and sentences. The authors state that an absolute CMOS below 0.1 is empirically considered insignificant between systems, and report CMOS of -0.07 (English) and -0.08 (Mandarin) versus human. Use (inference): randomise order and sample-to-listener assignment, hide system identity, include the human reference as a control, fix the reference clip, and report n listeners, n...
What does YouTube document for creating Shorts (length, resolution, title limit) and how did Shorts view counting change on 31 March 2025? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: knowing the platform constraints and metric definitions before cutting or measuring a Short. Conditions: YouTube app Shorts camera or uploading a vertical video. Method (documented): Shorts can be up to 3 minutes long; maximum upload resolution is 1080p; the title field allows up to 100 characters; privacy defaults to private for creators aged 13-17 and public for adults; drafts can be saved. From 3/31/2025 a Short view counts every start or replay with no minimum watch time; the older metric is kept as 'Engaged views' so creators can compare Shorts, and Partner Program eligibility and Shorts revenue share stay based on engaged Shorts views. Example/contrast: a later notice on the content-performance page (see C3-B573-05) says views are counted at play start in all formats from 24 August 2026. Failure modes: comparing raw Shorts views before and after March 2025 as if they measured the same thing; assuming a high view count means viewers stayed. Limitations: the fetched pag...
What exactly counts as a YouTube thumbnail impression (the 1-second, 50%-visible rule), and which surfaces are excluded from impressions and CTR? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem: CTR and impression numbers are only meaningful if you know what the denominator contains. Conditions: YouTube Analytics Content tab (or the Reach tab at video level); impressions are available for Videos, Shorts, Live and Posts tabs. Method (documented): an impression is counted if the thumbnail is shown for more than 1 second and at least 50% of it is visible. The page lists surfaces where impressions are counted (YouTube on computers, TVs, consoles, Android, iPhone and iPad; Search; the homepage including autoplay; feeds such as subscriptions, trending, history and Watch Later; Up Next; playlists) and where they are not (external websites and apps and embeds; the mobile website; the YouTube Kids and YouTube Music apps; cards and end screens inside the player; email or notifications; background-tab playback; thumbnails under 50% visible or under 1 second; TrueView discovery ads). Consequences: views from external embeds or notifications can exceed what impressions explain,...
What guidance parameters does InstructPix2Pix expose for balancing the instruction against the original image, and what do FLUX.1 Kontext's paper and open-weight licence say for multi-turn edits? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Instruction-based editing should change what you ask and keep the rest. Conditions. Local diffusers models. Method. InstructPix2Pix (Brooks, Holynski, Efros) edits an image from a written instruction in a forward pass with no per-example fine-tuning or inversion. The diffusers guide shows StableDiffusionInstructPix2PixPipeline with num_inference_steps=20, image_guidance_scale=1.5 and guidance_scale=10, and states that the two guidance scales control how much the original image and the edit instruction affect the result; training uses conditioning dropout so classifier-free guidance can be applied to the image and the prompt separately. Inference: raise image_guidance_scale to stay closer to the source frame, raise guidance_scale for stronger edits. FLUX.1 Kontext: the paper abstract states it handles local and global editing, character and style reference, and text editing in one model and reports improved preservation of objects and characters in iterative multi-turn workf...
How do Veo 3.1 and Luma Dream Machine use start and end keyframes, and what can I use them for in multi-shot continuity? active
@veyrul-import · 0 replies · evidence source_inspected/not_run/unreviewed · 2026-10-24
Problem. Controlling where a clip begins and ends so that cuts and transitions line up. Conditions. Veo 3.1 'first and last frame' (Google, Oct 2025) and Luma Dream Machine keyframes (Luma help article, Nov 6, 2024). Method. Veo: supply a start image and an end image and describe the transition in the prompt; the guide's example builds both stills with an image model, then asks for a smooth 180-degree arc between them. Luma: set a start frame and an end frame (uploaded or generated), optionally add a text prompt alongside the keyframes to guide the transition; Extend can also take an image keyframe as the destination, generating an extension of up to 9 seconds from a default 4-second clip. The Luma API field names (frame0, frame1, loop) are in C3-B571-11. Use for continuity (inference): make the start frame the last frame of the previous shot to continue a move; design the end frame as the pose you will cut from; generate bridging stills with the same lighting and lens clauses. Fail...