Research map / Label-efficient and robust learning / Self-supervised and contrastive
From Phase to Phenomenon: Self-Supervised Learning of Subsurface Scattering with Minimal Phase-shift Inputs
Arjun Majumdar, Raphael Braun, Andreas Engelhardt, Hendrik Lensch
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 PSP-specific augmentations instead of ImageNet augmentations (ablation) | SimSiam encoder · kNN material classification, Apple | kNN accuracy (%): 99.04 -> 99.6 | Improved |
| use PSP-specific augmentations instead of ImageNet augmentations (ablation) | SimSiam encoder · kNN material classification, Pear | kNN accuracy (%): 98.74 -> 99.99 | Improved |
| use PSP-specific augmentations instead of ImageNet augmentations (ablation) | SimSiam encoder · kNN material classification, Orange | kNN accuracy (%): 98.92 -> 99.99 | Improved |
| use PSP-specific augmentations instead of ImageNet augmentations (ablation) | SimSiam encoder · kNN material classification, Star | kNN accuracy (%): 99.93 -> 99.95 | No clear effect |
| use PSP-specific augmentations instead of ImageNet augmentations (ablation) | SimSiam encoder · kNN material classification, Shovel | kNN accuracy (%): 84.91 -> 91.43 | Improved |
| use SSL-pretrained encoder with 8 PSP images instead of capturing a footprint grid | relighting pipeline · image acquisition per new object | number of images: 3025 -> 8 | Improved |
| modify SimSiam hyperparameters (lr 0.05 without scaling, 20 warmup epochs, cosine decay) | SimSiam pretraining · PSP patches | dimensional collapse | Improved |
Closest papers by meaning
- Subsurface Scattering for Gaussian Splatting (NeurIPS 2024 · similarity 0.87)
- Decoupling Scattering: Pseudo-Label Guided NeRF for Scenes with Scattering Media (AAAI 2025 · similarity 0.85)
- WaveMo: Learning Wavefront Modulations to See Through Scattering (CVPR 2024 · similarity 0.84)
- OpticalNet: An Optical Imaging Dataset and Benchmark Beyond the Diffraction Limit (CVPR 2025 · similarity 0.84)
- Practical Measurements of Translucent Materials with Inter-Pixel Translucency Prior (CVPR 2024 · similarity 0.84)
- Light Transport-aware Diffusion Posterior Sampling for Single-View Reconstruction of 3D Volumes (CVPR 2025 · similarity 0.83)
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.
