Research map / 3D vision / 3D generation
View-Consistent 4D Light Field Superpixel Segmentation
Numair Khan, Qian Zhang, Lucas Kasser, Henry Stone, Min H. Kim, James Tompkin
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 occlusion-aware EPI segmentation with all-view propagation instead of LFSP central-view propagation | light field superpixels · HCI, estimated disparity | self similarity error and labels per pixel | Improved |
| use our method instead of LFSP with ground truth disparity | light field superpixels · HCI, ground truth disparity | self similarity error and labels per pixel | Improved |
| use our method instead of k-means with ground truth disparity | light field superpixels · HCI, ground truth disparity | number of labels per pixel | Hurt |
| use our method instead of k-means with ground truth disparity | light field superpixels · HCI, ground truth disparity | self similarity error | Improved |
| use our method instead of LFSP | light field superpixels · HCI, sizes 15-40 | achievable accuracy, boundary recall, undersegmentation error | Improved |
| use our method instead of k-means with ground truth disparity at larger superpixel sizes | light field superpixels · HCI, large sizes | accuracy and undersegmentation error | Hurt |
| use our method instead of k-means with ground truth disparity at small sizes (15-25) | light field superpixels · HCI, sizes 15-25 | accuracy and undersegmentation error | No clear effect |
| use our method instead of k-means | light field superpixels · HCI | boundary recall | Hurt |
| use our method instead of LFSP at small superpixel sizes (15-25) | light field superpixels · HCI, sizes 15-25 | compactness | No clear effect |
| use our method instead of LFSP at larger superpixel sizes | light field superpixels · HCI, large sizes | compactness | Improved |
| use our method instead of k-means | light field superpixels · HCI | compactness | Improved |
| use our method (3.3 minutes) instead of LFSP plus Wang disparity (about 10 minutes) | light field superpixels · HCI 9x9 768x768 | computation time: ~8 minutes + ~2 minutes -> ~3.3 minutes | Improved |
Closest papers by meaning
- 4D Light Field Superpixel and Segmentation (CVPR 2017 · similarity 0.89)
- Occlusion-Aware Method for Temporally Consistent Superpixels (TPAMI 2018 · similarity 0.87)
- Depth Estimation with Occlusion Modeling Using Light-Field Cameras (TPAMI 2016 · similarity 0.87)
- Epipolar Consistent Attention Aggregation Network for Unsupervised Light Field Disparity Estimation (ICCV 2025 · similarity 0.87)
- Learning Light Field Angular Super-Resolution via a Geometry-Aware Network (AAAI 2020 · similarity 0.86)
- Occlusion-Embedded Hybrid Transformer for Light Field Super-Resolution (AAAI 2025 · similarity 0.85)
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.
