Research map / Detection and segmentation / CNN features
Superpixel Segmentation With Fully Convolutional Networks
Check 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 conventional downsampling and bilinear upsampling with superpixel-based operations and jointly train | PSMNet · SceneFlow test | EPE: 1.04 -> 0.93 | Helped |
| jointly train the superpixel and stereo networks instead of fixing the superpixel network (ablation) | SP-PSMNet · SceneFlow test | EPE: 1.07 -> 0.93 | Helped |
| replace conventional downsampling and bilinear upsampling with superpixel-based operations and jointly train | PSMNet · HR-VS test | EPE: 3.83 -> 2.77 | Helped |
| jointly train the superpixel and stereo networks instead of fixing the superpixel network (ablation) | SP-PSMNet · HR-VS test | EPE: 3.70 -> 2.77 | Helped |
| replace conventional downsampling and bilinear upsampling with superpixel-based operations and jointly train | PSMNet · Middlebury-v3 | avgerr: 8.78 -> 7.11 | Helped |
| replace conventional downsampling and bilinear upsampling with superpixel-based operations and jointly train | PSMNet · Middlebury-v3 | rms: 23.3 -> 19.1 | Helped |
| replace conventional downsampling and bilinear upsampling with superpixel-based operations and jointly train | PSMNet · Middlebury-v3 | bad-4.0: 29.2 -> 27.5 | Helped |
| replace conventional downsampling and bilinear upsampling with superpixel-based operations and jointly train | PSMNet · Middlebury-v3 | A90: 22.8 -> 13.8 | Helped |
| use a fixed, pretrained superpixel network rather than jointly training it with stereo estimation (ablation) | SP-PSMNet · SceneFlow test | EPE: 1.04 -> 1.07 | Hurt |
Closest papers by meaning
- Learning Superpixels With Segmentation-Aware Affinity Loss (CVPR 2018 · similarity 0.91)
- Learning the Superpixel in a Non-Iterative and Lifelong Manner (CVPR 2021 · similarity 0.89)
- Computer Vision with a Superpixelation Camera (CVPR 2026 · similarity 0.89)
- Learning Invariant Inter-pixel Correlations for Superpixel Generation (AAAI 2024 · similarity 0.86)
- Hierarchical Histogram Threshold Segmentation - Auto-terminating High-detail Oversegmentation (CVPR 2024 · similarity 0.84)
- SuperPrimitive: Scene Reconstruction at a Primitive Level (CVPR 2024 · similarity 0.84)
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.