Research map / Video / Video generation
GraphVid: Interactive Graph-Controllable Video Generation
Vedant Shah, Onkar Susladkar, Tushar Prakash, Kiet Nguyen, Tianjiao (Joey) Yu, Adheesh Juvekar, Muntasir Wahed, Ismini Lourentzou
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use directed interaction graph conditioning instead of text+image (ablation) | GraphVid (LTX-Video) · MoveBench interaction subset | FID: 20.04 -> 17.02 | Improved |
| add semantic edge text to topology-only graph (ablation) | GraphVid · MoveBench interaction subset | FID: 17.52 -> 17.02 | Improved |
| add semantic edge text to topology-only graph (ablation) | GraphVid · MoveBench interaction subset | PSNR: 16.88 -> 15.98 | Hurt |
| add edge embeddings and message passing (ablation) | GraphVid · MoveBench interaction subset | FID: 19.55 -> 17.02 | Improved |
| increase LoRA rank from 16 to 128 (ablation) | GraphVid · MoveBench interaction subset | FID: 18.34 -> 17.02 | Improved |
| increase max node capacity from 30 to 256 (ablation) | GraphVid · MoveBench interaction subset | FVD: 99.42 -> 109.92 | Hurt |
| reduce max node capacity from 30 to 10 (ablation) | GraphVid · MoveBench interaction subset | FVD: 99.42 -> 102.04 | Hurt |
| swap LTX backbone for Wan 2.2 (ablation) | GraphVid · MoveBench interaction subset | FID: 17.02 -> 17.29 | No clear effect |
| graph conditioning vs WISA physics text | GraphVid · MoveBench interaction subset | FID: 25.98 -> 17.02 | Improved |
| graph conditioning vs Motion-I2V | GraphVid · MoveBench interaction subset | PSNR: 9.87 -> 15.98 | Improved |
| GraphVid vs Wan-Move (large-scale) | GraphVid · MoveBench interaction subset | FID: 15.56 vs 17.02 | Hurt |
| graph conditioning on multi-object scenes vs FlashMotion | GraphVid · MoveBench multi-object subset | FID: 55.03 -> 49.45 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.
- Wan-Move — Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance (NeurIPS 2025) · better on 1 of 9 shared results
- WISA — WISA: World simulator assistant for physics-aware text-to-video generation (NeurIPS 2025) · better on 5 of 5 shared results
- FlashMotion — FlashMotion: Few-Step Controllable Video Generation with Trajectory Guidance (CVPR 2026) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance (NeurIPS 2025 · similarity 0.89)
- IM-Zero: Instance-level Motion Controllable Video Generation in a Zero-shot Manner (CVPR 2025 · similarity 0.89)
- Motion Prompting: Controlling Video Generation with Motion Trajectories (CVPR 2025 · similarity 0.89)
- TrackGo: A Flexible and Efficient Method for Controllable Video Generation (AAAI 2025 · similarity 0.88)
- CAGE: Unsupervised Visual Composition and Animation for Controllable Video Generation (AAAI 2025 · similarity 0.88)
- Learn the Force We Can: Enabling Sparse Motion Control in Multi-Object Video Generation (AAAI 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
