Research map / Detection and segmentation / CNN features
Fast, Exact and Multi-scale Inference for Semantic Image Segmentation with Deep Gaussian CRFs
Siddhartha Chandra, Iasonas Kokkinos
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 8-connected instead of 4-connected neighbourhood (ablation) | QO · VOC12 val | mIoU: 76.36 -> 76.40 | No clear effect |
| use 12-connected instead of 4-connected neighbourhood (ablation) | QO · VOC12 val | mIoU: 76.36 -> 76.42 | No clear effect |
| use shared Potts-type pairwise terms (ablation) | QO · VOC12 val | mIoU: 76.36 -> 76.59 | Improved |
| add one QO unit per image resolution (ablation) | QO · VOC12 val | mIoU: 76.36 -> 76.69 | Improved |
| couple scales with joint multi-resolution QO instead of decoupled per-resolution QO (ablation) | QO^res · VOC12 val | mIoU: 76.69 -> 76.93 | Improved |
| add QO module to basenet | Basenet · VOC12 test | mIoU: 72.72 -> 73.41 | Improved |
| add multi-resolution QO to basenet | Basenet · VOC12 test | mIoU: 72.72 -> 73.86 | Improved |
| add DenseCRF post-processing (ablation) | QO^mres · VOC12 test | mIoU: 73.86 -> 75.46 | Improved |
| add DenseCRF post-processing (ablation) | Basenet · VOC12 test | mIoU: 72.72 -> 73.78 | Improved |
| add CRF post-processing to Potts QO on Resnet-101 (ablation) | QO^s Deeplab-v2 Resnet-101 · VOC12 test | mIoU: 79.5 -> 80.2 | Improved |
| use QO^s instead of Deeplab-v2 with CRF, without CRF | Deeplab-v2 Resnet-101 · VOC12 test | mIoU: 79.7 -> 79.5 | No clear effect |
| replace parallel mean-field (Jacobi) with conjugate gradient solver | QO inference · GPU inference speed | speed: up to 2.5 faster | Improved |
Closest papers by meaning
- Dense and Low-Rank Gaussian CRFs Using Deep Embeddings (ICCV 2017 · similarity 0.93)
- Gaussian Conditional Random Field Network for Semantic Segmentation (CVPR 2016 · similarity 0.92)
- Higher Order Conditional Random Fields in Deep Neural Networks (ECCV 2016 · similarity 0.91)
- Efficient Linear Programming for Dense CRFs (CVPR 2017 · similarity 0.91)
- Regularized Frank-Wolfe for Dense CRFs: Generalizing Mean Field and Beyond (NeurIPS 2021 · similarity 0.90)
- Efficient Piecewise Training of Deep Structured Models for Semantic Segmentation (CVPR 2016 · similarity 0.90)
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.
