Research map / Detection and segmentation / Open-set and few-shot detection
What Is Around the Camera?
Stamatios Georgoulis, Konstantinos Rematas, Tobias Ritschel, Mario Fritz, Tinne Tuytelaars, Luc Van Gool
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 |
|---|---|---|---|
| add background network (ablation) | Singlet · Synthetic Cars (single material) | DSSIM: .311 -> .281 | Improved |
| add background network (ablation) | Singlet · Real Spheres | DSSIM: .324 -> .360 | Hurt |
| add background network (ablation) | Singlet · Real Cars (single material) | DSSIM: .337 -> .360 | Hurt |
| add background network (ablation) | Triplets · Synthetic Cars (single material) | DSSIM: .268 -> .210 | Improved |
| add background network (ablation) | Triplets · Synthetic Cars (multi-material) | DSSIM: .277 -> .226 | Improved |
| add background network (ablation) | Triplets · Real Cars (single material) | DSSIM: .332 -> .315 | Improved |
| add background network (ablation) | Triplets · Real Cars (multi-material) | DSSIM: .284 -> .272 | Improved |
| use three reflectance maps instead of one (ablation) | Singlet · Synthetic Cars (single material) | DSSIM: .311 -> .268 | Improved |
| learned fusion of three maps instead of mask-aware mean of singlet predictions (ablation) | Mask-aware mean · Real Cars (multi-material) | DSSIM: .305 -> .284 | Improved |
| increase number of reflectance maps from 3 to 5 (no retraining) (ablation) | Triplets + BG · Real Spheres | DSSIM: .305 -> .292 | Improved |
| increase number of reflectance maps from 3 to 5 (no retraining) (ablation) | Triplets + BG · Real Cars (single material) | DSSIM: .315 -> .295 | Improved |
| increase number of reflectance maps from 3 to 4 (no retraining) (ablation) | Triplets + BG · Real Spheres | DSSIM: .305 -> .309 | Hurt |
Closest papers by meaning
- Reflectance and Illumination Recovery in the Wild (TPAMI 2016 · similarity 0.87)
- Eclipse: Disambiguating Illumination and Materials using Unintended Shadows (CVPR 2024 · similarity 0.83)
- SfPUEL: Shape from Polarization under Unknown Environment Light (NeurIPS 2024 · similarity 0.83)
- MaterialRefGS: Reflective Gaussian Splatting with Multi-view Consistent Material Inference (NeurIPS 2025 · similarity 0.83)
- Diffusion Reflectance Map: Single-Image Stochastic Inverse Rendering of Illumination and Reflectance (CVPR 2024 · similarity 0.82)
- Factored-NeuS: Reconstructing Surfaces, Illumination, and Materials of Possibly Glossy Objects (CVPR 2025 · similarity 0.82)
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.
