How do MotionCtrl and CameraCtrl differ from text camera prompts: what do they condition on, what can be controlled independently, and which backbones are supported?

Problem. Words like 'dolly in' give no exact trajectory. Pose-conditioned models take a numeric camera path. Conditions. Research checkpoints you run yourself (not hosted prompt UIs). Method. MotionCtrl: motion conditions are camera poses and object trajectories, treated as appearance-free; camera motion and object motion can be controlled independently or together; the repo script sets condtype to 'camera_motion', 'object_motion' or 'both'; released on VideoCrafter, SVD and AnimateDiff branches; extra test poses in dataset/camera_poses; object trajectories can be drawn with HandyTrajDrawer or extracted with ParticleSfM. CameraCtrl: a plug-and-play camera pose control module trained on top of a video diffusion model, leaving other base modules untouched; main branch is on AnimateDiff V3 with an svd branch; trajectories are per-frame camera poses (the README visualises each frame as a small tetrahedron for camera position and orientation); paper reports that training videos with diverse camera distributions and appearance similar to the base model improve controllability and generalisation. Craft use (inference): a pan or tilt is a rotation-only pose sequence, a dolly/truck/pedestal is translation-only, a crane combines translation with rotation, and zoom (focal length) is not obviously covered by pose-only conditioning since pose files carry extrinsics, so it is likely fixed by the training intrinsics. Failure modes. Not stated in the sources fetched beyond the training-data finding; apparent pose-range limits follow dataset camera distribution (RealEstate10K-style indoor tours are the cited trajectory source in the README data prep). Limitations. Backbones are 2023-24 models with short low-resolution clips. Version/date. MotionCtrl SIGGRAPH 2024 (arXiv 2312.03641); CameraCtrl arXiv 2404.02101. Curator qualifications: All cited documentation excerpts located on the live pages by curator GET; no video sources. Limitations: Zoom/intrinsics point is inference: I did not read the CameraCtrl or MotionCtrl code for focal-length handling. CameraCtrl licence file was not found at the raw path tried. Evidence status: statements located in the cited sources by a correlated desk review (Claude); not executed; not independently reviewed.

{
  "citations": [
    {
      "url": "https://arxiv.org/abs/2312.03641",
      "note": "Abstract; Inference; Release notes; Abstract; top of README; visualisation note; mapping to camera grammar"
    },
    {
      "url": "https://raw.githubusercontent.com/TencentARC/MotionCtrl/main/README.md",
      "note": "Abstract; Inference; Release notes; Abstract; top of README; visualisation note; mapping to camera grammar"
    },
    {
      "url": "https://arxiv.org/abs/2404.02101",
      "note": "Abstract; Inference; Release notes; Abstract; top of README; visualisation note; mapping to camera grammar"
    },
    {
      "url": "https://raw.githubusercontent.com/hehao13/CameraCtrl/main/README.md",
      "note": "Abstract; Inference; Release notes; Abstract; top of README; visualisation note; mapping to camera grammar"
    }
  ]
}