DSER: Spectral Epipolar Representation for Efficient Light Field Depth Estimation
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911572757053440 |
|---|---|
| author | Mohammad, Noor Islam S. Meherab, Md Muntaqim |
| author_facet | Mohammad, Noor Islam S. Meherab, Md Muntaqim |
| contents | Dense light field depth estimation remains challenging due to sparse angular sampling, occlusion boundaries, textureless regions, and the cost of exhaustive multi-view matching. We propose \emph{Deep Spectral Epipolar Representation} (DSER), a geometry-aware framework that introduces spectral regularization in the epipolar domain for dense disparity reconstruction. DSER models frequency-consistent EPI structure to constrain correspondence estimation and couples this prior with a hybrid inference pipeline that combines least squares gradient initialization, plane-sweeping cost aggregation, and multiscale EPI refinement. An occlusion-aware directed random walk further propagates reliable disparity along edge-consistent paths, improving boundary sharpness and weak-texture stability. Experiments on benchmark and real-world light field datasets show that DSER achieves a strong accuracy-efficiency trade-off, producing more structurally consistent depth maps than representative classical and hybrid baselines. These results establish spectral epipolar regularization as an effective inductive bias for scalable and noise-robust light field depth estimation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_08900 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DSER: Spectral Epipolar Representation for Efficient Light Field Depth Estimation Mohammad, Noor Islam S. Meherab, Md Muntaqim Computer Vision and Pattern Recognition 68T45, 68U10 I.4.8; I.2.10 Dense light field depth estimation remains challenging due to sparse angular sampling, occlusion boundaries, textureless regions, and the cost of exhaustive multi-view matching. We propose \emph{Deep Spectral Epipolar Representation} (DSER), a geometry-aware framework that introduces spectral regularization in the epipolar domain for dense disparity reconstruction. DSER models frequency-consistent EPI structure to constrain correspondence estimation and couples this prior with a hybrid inference pipeline that combines least squares gradient initialization, plane-sweeping cost aggregation, and multiscale EPI refinement. An occlusion-aware directed random walk further propagates reliable disparity along edge-consistent paths, improving boundary sharpness and weak-texture stability. Experiments on benchmark and real-world light field datasets show that DSER achieves a strong accuracy-efficiency trade-off, producing more structurally consistent depth maps than representative classical and hybrid baselines. These results establish spectral epipolar regularization as an effective inductive bias for scalable and noise-robust light field depth estimation. |
| title | DSER: Spectral Epipolar Representation for Efficient Light Field Depth Estimation |
| topic | Computer Vision and Pattern Recognition 68T45, 68U10 I.4.8; I.2.10 |
| url | https://arxiv.org/abs/2508.08900 |