Research map / Detection and segmentation / CNN features
CCNet: Criss-Cross Attention for Semantic Segmentation
Zilong Huang, Xinggang Wang, Lichao Huang, Chang Huang, Yunchao Wei, Wenyu Liu
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 |
|---|---|---|---|
| add one criss-cross attention loop (R=1) (ablation) | ResNet-101 FCN baseline · Cityscapes val | mIoU (%): 75.1 -> 78.0 | Improved |
| increase RCCA loops from 1 to 2 (ablation) | ResNet-101 RCCA · Cityscapes val | mIoU (%): 78.0 -> 79.8 | Improved |
| increase RCCA loops from 2 to 3 (ablation) | ResNet-101 RCCA · Cityscapes val | mIoU (%): 79.8 -> 80.2 | Improved |
| add category consistent loss (CCL) (ablation) | ResNet-101 RCCA(R=2) · Cityscapes val | mIoU (%): 79.8 -> 80.5 | Improved |
| add category consistent loss (CCL) (ablation) | ResNet-50 RCCA(R=2) · Cityscapes val | mIoU (%): 78.5 -> 79.3 | Improved |
| use piece-wise instead of quadratic distance in CCL (ablation) | ResNet-50 RCCA+CCL · Cityscapes val | training success rate: 6/10 -> 9/10 | Improved |
| use piece-wise instead of quadratic distance in CCL (ablation) | ResNet-50 RCCA+CCL · Cityscapes val | mIoU (%): 79.2 -> 79.3 | No clear effect |
| use RCCA(R=2) instead of non-local block (ablation) | ResNet-50 · Cityscapes val | mIoU (%): 77.3 -> 78.5 | Improved |
| use RCCA instead of non-local in terms of memory (ablation) | ResNet-50 · Cityscapes val | memory increment (M): 1411 -> 127 | Improved |
| run non-local module recurrently (R=2) (ablation) | ResNet-50 +NL · Cityscapes val | mIoU (%): 77.3 -> 78.7 | Improved |
| use full criss-cross shape instead of stacked horizontal-vertical attention (ablation) | ResNet-50 · Cityscapes val | mIoU (%): 77.3 -> 78.5 | Improved |
| increase input frames from T=1 to T=5 in 3D criss-cross attention (ablation) | CCNet3D · CamVid test | mIoU (%): 77.9 -> 79.1 | Improved |
Later papers that compared with this method
16 later accepted papers report CCNet as a baseline in their results tables; 13 of them report a better number on most of the results they share with it.
- Hilbert Curve-Based Attention Enabling Topology-Preserving Image Tensor Representation for Semantic Segmentation Network (CVPR 2026 · better on 2 of 2 shared results)
- Anomaly Detection of Integrated Circuits Package Substrates Using the Large Vision Model SAIC: Dataset Construction, Methodology, and Application (ICCV 2025 · better on 2 of 2 shared results)
- TOP-ReID: Multi-Spectral Object Re-identification with Token Permutation (AAAI 2024 · better on 1 of 3 shared results)
- Vision Transformers: From Semantic Segmentation to Dense Prediction (IJCV 2024 · better on 0 of 1 shared result)
- Multispectral Video Semantic Segmentation: A Benchmark Dataset and Baseline (CVPR 2023 · better on 2 of 2 shared results)
- Large-Field Contextual Feature Learning for Glass Detection (TPAMI 2022 · better on 2 of 2 shared results)
- Sparse and Complete Latent Organization for Geospatial Semantic Segmentation (CVPR 2022 · better on 1 of 1 shared result)
- Rethinking Semantic Segmentation From a Sequence-to-Sequence Perspective With Transformers (CVPR 2021 · better on 1 of 2 shared results)
8 more not listed.
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PSANet — PSANet: Point-wise Spatial Attention Network for Scene Parsing (ECCV 2018) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs (TPAMI 2017 · similarity 0.87)
- CLIMS: Cross Language Image Matching for Weakly Supervised Semantic Segmentation (CVPR 2022 · similarity 0.87)
- ACFNet: Attentional Class Feature Network for Semantic Segmentation (ICCV 2019 · similarity 0.87)
- Memory-Based Cross-Image Contexts for Weakly Supervised Semantic Segmentation (TPAMI 2022 · similarity 0.87)
- Mining Cross-Image Semantics for Weakly Supervised Semantic Segmentation (ECCV 2020 · similarity 0.87)
- ResCLIP: Residual Attention for Training-free Dense Vision-language Inference (CVPR 2025 · similarity 0.86)
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.
