GeCo: Evaluating Geometric Consistency for Video Generation via Motion and Structure
Not on the map. Why?
Abstract (arXiv)
We introduce GeCo, a geometry-grounded metric for jointly detecting geometric deformation and occlusion-inconsistency artifacts in static scenes. By fusing residual motion and depth priors, GeCo produces interpretable, dense consistency maps that reveal these artifacts. We use GeCo to systematically benchmark recent video generation models, uncovering common failure modes, and further employ it as a training-free guidance loss to reduce deformation artifacts during video generation.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 50 of its 60 references that are in the map. First posted on arXiv 25 Dec 2025.
Builds on
50 of 60 references.
References · 50
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
- 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
- 2023Adding Conditional Control to Text-to-Image Diffusion Modelscited by 5,466
- 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
- 2024SAM 2: Segment Anything in Images and Videoscited by 2,784
- 2025Wan: Open and Advanced Large-Scale Video Generative Modelscited by 2,536
- 2024CogVideoX: Text-to-Video Diffusion Models with An Expert Transformercited by 2,198
- 2023VBench: Comprehensive Benchmark Suite for Video Generative Modelscited by 1,755
- 2024HunyuanVideo: A Systematic Framework For Large Video Generative Modelscited by 1,597
- 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,526
- 2025VGGT: Visual Geometry Grounded Transformercited by 1,475
- 2022RePaint: Inpainting using Denoising Diffusion Probabilistic Modelscited by 1,419
38 more
- 2024Depth Anything V2cited by 1,385
- 2017SuperPoint: Self-Supervised Interest Point Detection and Descriptioncited by 1,313
- 2022CogVideo: Large-scale Pretraining for Text-to-Video Generation via Transformerscited by 973
- 2025Cosmos World Foundation Model Platform for Physical AIcited by 833
- 2023ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenescited by 747
- 2021DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Camerascited by 673
- 2021Common Objects in 3D: Large-Scale Learning and Evaluation of Real-life 3D Category Reconstructioncited by 646
- 2023ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depthcited by 646
- 2023LightGlue: Local Feature Matching at Light Speedcited by 583
- 2022Zero-Shot Image Restoration Using Denoising Diffusion Null-Space Modelcited by 574
- 2024LTX-Video: Realtime Video Latent Diffusioncited by 549
- 2023DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Visioncited by 514
- 2020TartanAir: A Dataset to Push the Limits of Visual SLAMcited by 494
- 2023Universal Guidance for Diffusion Modelscited by 426
- 2024SV3D: Novel Multi-view Synthesis and 3D Generation from a Single Image using Latent Video Diffusioncited by 364
- 2023EvalCrafter: Benchmarking and Evaluating Large Video Generation Modelscited by 332
- 2025GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Controlcited by 325
- 2023ControlVideo: Training-free Controllable Text-to-Video Generationcited by 324
- 2025Matrix-game 2.0: An open-source real-time and streaming interactive world modelcited by 174
- 2024Towards World Simulator: Crafting Physical Commonsense-Based Benchmark for Video Generationcited by 167
- 2023Manifold Preserving Guided Diffusioncited by 153
- 2025WorldPlay: Towards Long-Term Geometric Consistency for Real-Time Interactive World Modelingcited by 152
- 2025WorldScore: A Unified Evaluation Benchmark for World Generationcited by 143
- 2025VideoPhy-2: A Challenging Action-Centric Physical Commonsense Evaluation in Video Generationcited by 141
- 2025MEt3R: Measuring Multi-View Consistency in Generated Imagescited by 91
- 2025RELIC: Interactive Video World Model with Long-Horizon Memorycited by 89
- 2024PhyT2V: LLM-Guided Iterative Self-Refinement for Physics-Grounded Text-to-Video Generationcited by 77
- 2023Long-Term Photometric Consistent Novel View Synthesis with Diffusion Modelscited by 76
- 2024World-consistent Video Diffusion with Explicit 3D Modelingcited by 62
- 2024VFusion3D: Learning Scalable 3D Generative Models from Video Diffusion Modelscited by 56
- 2024Motion Guidance: Diffusion-Based Image Editing with Differentiable Motion Estimatorscited by 39
- 2025UFM: A Simple Path towards Unified Dense Correspondence with Flowcited by 35
- 2025Segment Any Motion in Videoscited by 34
- 2025VideoScene: Distilling Video Diffusion Model to Generate 3D Scenes in One Stepcited by 27
- 2024Optical-Flow Guided Prompt Optimization for Coherent Video Generationcited by 17
- 2025Frame Guidance: Training-Free Guidance for Frame-Level Control in Video Diffusion Modelscited by 11
- 2025JOG3R: Towards 3D-Consistent Video Generatorscited by 8
- 2024Multistable Shape from Shading Emerges from Patch Diffusioncited by 5
Built on by
Citing papers per year: 2026: 1.
1 paper.