Research map / Detection and segmentation / CNN features
Memory-oriented Decoder for Light Field Salient Object Detection
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| replace RGB input with light field focal stack (ablation) | baseline network · Ours dataset | F-measure: 0.643 -> 0.805 | Helped |
| weight focal slice features by attention before concatenation (ablation) | baseline LF · Ours dataset | MAE: 0.074 -> 0.069 | Helped |
| replace concatenation with ConvLSTM fusion (ablation) | LF with weighting · Ours dataset | MAE: 0.069 -> 0.062 | Helped |
| add global perception module (GPM) (ablation) | Mo-SFM · Ours dataset | MAE: 0.062 -> 0.059 | Helped |
| add Mo-FIM without SCIM (ablation) | SFM with GPM · Ours dataset | MAE: 0.059 -> 0.054 | Helped |
| add scene context integration module (SCIM) to Mo-FIM (ablation) | Mo-FIM · Ours dataset | MAE: 0.054 -> 0.052 | Helped |
| add SCIM to Mo-FIM (ablation) | Mo-FIM · LFSD | MAE: 0.092 -> 0.089 | Helped |
| use real focal slices instead of repeated RGB frames (ablation) | MoLF · LFSD | F-measure / MAE: 0.740 / 0.140 -> 0.819 / 0.089 | Helped |
| add Mo-SFM overall to baseline (ablation) | baseline LF · LFSD | MAE: 0.121 -> 0.099 | Helped |
Closest papers by meaning
- Learning From Pixel-Level Noisy Label: A New Perspective for Light Field Saliency Detection (CVPR 2022 · similarity 0.92)
- Exploit and Replace: An Asymmetrical Two-Stream Architecture for Versatile Light Field Saliency Detection (AAAI 2020 · similarity 0.90)
- Light Field Intrinsics With a Deep Encoder-Decoder Network (CVPR 2018 · similarity 0.88)
- Cascaded Partial Decoder for Fast and Accurate Salient Object Detection (CVPR 2019 · similarity 0.86)
- EPINET: A Fully-Convolutional Neural Network Using Epipolar Geometry for Depth From Light Field Images (CVPR 2018 · similarity 0.86)
- NeLF-Pro: Neural Light Field Probes for Multi-Scale Novel View Synthesis (CVPR 2024 · 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.