Pith. sign in

REVIEW 8 cited by

FreeTraj: Tuning-Free Trajectory Control in Video Diffusion Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2406.16863 v1 pith:Z4TYGHSZ submitted 2024-06-24 cs.CV

classification cs.CV
keywords trajectoryvideocontroldiffusiongenerationfreetrajgeneratedtrajectories
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Diffusion model has demonstrated remarkable capability in video generation, which further sparks interest in introducing trajectory control into the generation process. While existing works mainly focus on training-based methods (e.g., conditional adapter), we argue that diffusion model itself allows decent control over the generated content without requiring any training. In this study, we introduce a tuning-free framework to achieve trajectory-controllable video generation, by imposing guidance on both noise construction and attention computation. Specifically, 1) we first show several instructive phenomenons and analyze how initial noises influence the motion trajectory of generated content. 2) Subsequently, we propose FreeTraj, a tuning-free approach that enables trajectory control by modifying noise sampling and attention mechanisms. 3) Furthermore, we extend FreeTraj to facilitate longer and larger video generation with controllable trajectories. Equipped with these designs, users have the flexibility to provide trajectories manually or opt for trajectories automatically generated by the LLM trajectory planner. Extensive experiments validate the efficacy of our approach in enhancing the trajectory controllability of video diffusion models.

Discussion (0). Sign in to comment.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Robot-Factored World Models via Robot Rendering

    cs.RO 2026-07 conditional novelty 6.0 of 10

    Conditioning a video world model on rendered nominal robot trajectories (URDF mesh + depth) instead of raw actions or logged future states improves action-following and enables zero-shot embodiment change.

  2. SymphoMotion: Joint Control of Camera Motion and Object Dynamics for Coherent Video Generation

    cs.CV 2026-04 conditional novelty 6.0 of 10

    SymphoMotion jointly controls camera trajectories and depth-aware object dynamics inside one video diffusion model, supported by the new RealCOD-25K real-world paired-motion dataset.

  3. Streaming Drag-Oriented Interactive Video Manipulation: Drag Anything, Anytime!

    cs.CV 2025-10 conditional novelty 6.0 of 10

    DragStream enables real-time drag, deform, and rotate edits on autoregressively generated videos without retraining, by correcting latent drift and selectively filtering context features.

  4. AnyI2V: Animating Any Conditional Image with Motion Control

    cs.CV 2025-07 conditional novelty 6.0 of 10

    AnyI2V animates arbitrary conditional images with user-defined trajectories by injecting debiased diffusion features and aligning attention queries across frames, without training.

  5. When Distillation Breaks Motion Control: Restoring Generative Trajectories for Fast Video Generators

    cs.CV 2025-06 conditional novelty 6.0 of 10

    MotionEcho adaptively re-injects teacher-model guidance into few-step distilled video generators so reference motion can be copied at test time without training.

  6. Interactive Video Generation via Domain Adaptation

    cs.CV 2025-05 conditional novelty 6.0 of 10

    A training-free method combines mask normalization and temporal intrinsic denoising to improve trajectory control and perceptual quality in text-to-video diffusion.

  7. MOVi: Training-free Text-conditioned Multi-Object Video Generation

    cs.CV 2025-05 conditional novelty 6.0 of 10

    MOVi improves multi-object video generation without retraining by using LLM-planned trajectories to reinitialize the diffusion noise and by reweighting attention to stop objects from mixing together.

  8. From World Action Models to Embodied Brains: A Roadmap for Open-World Physical Intelligence

    cs.RO 2026-07 conditional novelty 4.0 of 10

    Physical intelligence needs an embodied brain that reasons over interventions and emits capability requests, grounded by a physical harness and shared experience contracts rather than direct actuator policies.

Pith tools