Research map / 3D vision / Depth and stereo
What You Can Learn by Staring at a Blank Wall
Prafull Sharma, Miika Aittala, Yoav Y. Schechner, Antonio Torralba, Gregory W. Wornell, William T. Freeman, Frédo Durand
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 |
|---|---|---|---|
| increase input segment from 64 to 256 frames (ablation) | Number-of-people CNN · Number-of-people CNN (test scenes) | accuracy (%): 79.4% -> 94.4% | Improved |
| increase input segment from 64 to 256 frames (ablation) | Activity recognition CNN · Activity recognition CNN (test scenes) | accuracy (%): 88% -> 93.7% | Improved |
| increase input segment from 128 to 256 frames (ablation) | Number-of-people CNN · Number-of-people classification (test scenes) | accuracy (%): 88.3% -> 94.4% | Improved |
| increase input segment from 128 to 256 frames (ablation) | Activity recognition CNN · Activity recognition (test scenes) | accuracy (%): 90.8% -> 93.7% | Improved |
| two people move with equal amplitude and in lockstep rather than differing motion (ablation) | Number-of-people CNN on synthetic data · Synthetic two-person segments | classification accuracy | Hurt |
| increase relative motion frequency of the second person (ablation) | Number-of-people CNN on synthetic data · Synthetic two-person segments | classification accuracy | Improved |
| make the two people's motion amplitudes differ (ablation) | Number-of-people CNN on synthetic data · Synthetic two-person segments | classification accuracy | Improved |
Closest papers by meaning
- Making the Invisible Visible: Action Recognition Through Walls and Occlusions (ICCV 2019 · similarity 0.85)
- Similarity-Consistent Likelihood Diffusion enables Hidden Person Detection from Wall Reflections (CVPR 2026 · similarity 0.84)
- Turning Corners Into Cameras: Principles and Methods (ICCV 2017 · similarity 0.84)
- Propagate and Calibrate: Real-Time Passive Non-Line-of-Sight Tracking (CVPR 2023 · similarity 0.83)
- Believing is Seeing: Unobserved Object Detection using Generative Models (CVPR 2025 · similarity 0.81)
- Can NeRFs "See" without Cameras? (NeurIPS 2025 · similarity 0.80)
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.
