Research map / Vision-language and multimodal / Human and embodied modeling
3DB: A Framework for Debugging Computer Vision Models
Guillaume Leclerc, Hadi Salman, Andrew Ilyas, Sai Vemprala, Logan Engstrom, Vibhav Vineet, Kai Yuanqing Xiao, Pengchuan Zhang, Shibani Santurkar, Greg Yang, Ashish Kapoor, Aleksander Madry
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 |
|---|---|---|---|
| render objects on varied HDRI backgrounds | ResNet-18 · 3DB simulated scenes | accuracy | Mixed |
| replace object texture with animal-skin texture | ResNet-18 · 3DB cue-conflict objects | accuracy on previously correct renders: drop 90-95% | Hurt |
| vary liquid inside mug | ResNet-18 · 3DB mug interior viewpoints | prediction stability: 20.7% of viewpoints keep constant prediction | Hurt |
| increase background complexity (edge density) | ResNet-18 · 3DB coffee mug | accuracy | Hurt |
| change object orientation or scale | ResNet-18 · 3DB renders | accuracy | Mixed |
| compose zoom and background controls | ResNet-18 · 3DB orange | accuracy | Mixed |
| simulated vs physical-world renderings | ResNet-18 · 3DB physical realism | agreement of correctness: 85% agreement | Improved |
Closest papers by meaning
- 3DCodeBench: Benchmarking Agentic Procedural 3D Modeling Via Code (NeurIPS 2026 · similarity 0.85)
- GIQ: Benchmarking 3D Geometric Reasoning of Vision Foundation Models with Simulated and Real Polyhedra (ICLR 2026 · similarity 0.84)
- HiBug2: Efficient and Interpretable Error Slice Discovery for Comprehensive Model Debugging (ICLR 2025 · similarity 0.84)
- BEHAVIOR Vision Suite: Customizable Dataset Generation via Simulation (CVPR 2024 · similarity 0.84)
- Evaluating Multiview Object Consistency in Humans and Image Models (NeurIPS 2024 · similarity 0.84)
- InfiniBench: Infinite Benchmarking for Visual Spatial Reasoning with Customizable Scene Complexity (CVPR 2026 · similarity 0.84)
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.
