Papers › SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation
SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation
Meng-Hao Guo, Cheng-Ze Lu, Qibin Hou, ZhengNing Liu, Ming-Ming Cheng, Shi-Min Hu
We present SegNeXt, a simple convolutional network architecture for semantic segmentation. Recent transformer-based models have dominated the field of semantic segmentation due to the efficiency of self-attention in encoding spatial information. In this paper, we show that convolutional attention is a more efficient and effective way to encode contextual information than the self-attention mechanism in transformers. By re-examining the characteristics owned by successful segmentation models, we discover several key components leading to the performance improvement of segmentation models. This motivates us to design a novel convolutional attention network that uses cheap convolutional operations. Without bells and whistles, our SegNeXt significantly improves the performance of previous state-of-the-art methods on popular benchmarks, including ADE20K, Cityscapes, COCO-Stuff, Pascal VOC, Pascal Context, and iSAID. Notably, SegNeXt outperforms EfficientNet-L2 w/ NAS-FPN and achieves 90.6% mIoU on the Pascal VOC 2012 test leaderboard using only 1/10 parameters of it. On average, SegNeXt achieves about 2.0% mIoU improvements compared to the state-of-the-art methods on the ADE20K datasets with the same or fewer computations. Code is available at https://github.com/uyzhang/JSeg (Jittor) and https://github.com/Visual-Attention-Network/SegNeXt (Pytorch).
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Real-Time Semantic Segmentation | Cityscapes val | SegNext-T-Seg100 | Frame (fps) | 28.1 | #4 of 24 | Archive leaderboard | report |
| Real-Time Semantic Segmentation | Cityscapes val | SegNext-T-Seg100 | mIoU | 79.8% | #4 of 24 | Archive leaderboard | report |
| Semantic Segmentation | DDD17 | SegNeXt-B | mIoU | 71.46 | #4 of 9 | Archive leaderboard | report |
| Semantic Segmentation | DSEC | SegNeXt-B | mIoU | 71.55 | #5 of 9 | Archive leaderboard | report |
| Semantic Segmentation | iSAID | SegNeXt-L | mIoU | 70.3 | #1 of 19 | Archive leaderboard | report |
| Semantic Segmentation | iSAID | SegNeXt-B | mIoU | 69.9 | #2 of 19 | Archive leaderboard | report |
| Semantic Segmentation | iSAID | SegNeXt-S | mIoU | 68.8 | #4 of 19 | Archive leaderboard | report |
| Semantic Segmentation | iSAID | SegNeXt-T | mIoU | 68.3 | #6 of 19 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections