Papers › Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes

Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes

15 Jan 2021arXiv:2101.06085archive 2025-07-28

Yuanduo Hong, Huihui Pan, Weichao Sun, Yisong Jia

Semantic segmentation is a key technology for autonomous vehicles to understand the surrounding scenes. The appealing performances of contemporary models usually come at the expense of heavy computations and lengthy inference time, which is intolerable for self-driving. Using light-weight architectures (encoder-decoder or two-pathway) or reasoning on low-resolution images, recent methods realize very fast scene parsing, even running at more than 100 FPS on a single 1080Ti GPU. However, there is still a significant gap in performance between these real-time methods and the models based on dilation backbones. To tackle this problem, we proposed a family of efficient backbones specially designed for real-time semantic segmentation. The proposed deep dual-resolution networks (DDRNets) are composed of two deep branches between which multiple bilateral fusions are performed. Additionally, we design a new contextual information extractor named Deep Aggregation Pyramid Pooling Module (DAPPM) to enlarge effective receptive fields and fuse multi-scale context based on low-resolution feature maps. Our method achieves a new state-of-the-art trade-off between accuracy and speed on both Cityscapes and CamVid dataset. In particular, on a single 2080Ti GPU, DDRNet-23-slim yields 77.4% mIoU at 102 FPS on Cityscapes test set and 74.7% mIoU at 230 FPS on CamVid test set. With widely used test augmentation, our method is superior to most state-of-the-art models and requires much less computation. Codes and trained models are available online.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

ydhongHIT/DDRNet officialmentioned on GitHubpytorch report
Deci-AI/super-gradients mentioned on GitHubpytorch report
hamidriasat/DDRNets mentioned on GitHubtf report
sithu31296/semantic-segmentation mentioned on GitHubpytorch report

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

All-day Semantic SegmentationAutonomous VehiclesReal-Time Semantic SegmentationScene ParsingSemantic Segmentation

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
All-day Semantic Segmentation All-day CityScapes DDR-Net mIoU 68.6 #2 of 3 Archive leaderboard report
Real-Time Semantic Segmentation CamVid DDRNet-23(Cityscapes-Pretrained) Frame (fps) 94(2080Ti) #4 of 29 Archive leaderboard report
Real-Time Semantic Segmentation CamVid DDRNet-23(Cityscapes-Pretrained) Time (ms) 10.6 #4 of 29 Archive leaderboard report
Real-Time Semantic Segmentation CamVid DDRNet-23(Cityscapes-Pretrained) mIoU 80.6 #4 of 29 Archive leaderboard report
Real-Time Semantic Segmentation CamVid DDRNet-23-slim Frame (fps) 230(2080Ti) #13 of 29 Archive leaderboard report
Real-Time Semantic Segmentation CamVid DDRNet-23-slim Time (ms) 4.3 #13 of 29 Archive leaderboard report
Real-Time Semantic Segmentation CamVid DDRNet-23-slim mIoU 74.7 #13 of 29 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test DDRNet-23-slim Frame (fps) 101.6(2080Ti) #7 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test DDRNet-23-slim Time (ms) 9.8 #7 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes test DDRNet-23-slim mIoU 77.4% #7 of 39 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes val DDRNet23 Frame (fps) 37.1 #5 of 24 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes val DDRNet23 mIoU 79.4 #5 of 24 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes val DDRNet23-slim Frame (fps) 101.6 #10 of 24 Archive leaderboard report
Real-Time Semantic Segmentation Cityscapes val DDRNet23-slim mIoU 77.4 #10 of 24 Archive leaderboard report
Semantic Segmentation CamVid DDRNet23 Mean IoU 80.6% #7 of 21 Archive leaderboard report
Semantic Segmentation Cityscapes test DDRNet-39 1.5x Mean IoU (class) 82.4% #28 of 105 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

Average PoolingBatch NormalizationConvolutionPyramid Pooling ModuleReLU

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