HyperAI超神経
ホームニュース最新論文チュートリアルデータセット百科事典SOTALLMモデルGPU ランキング学会
検索
サイトについて
HyperAI超神経

Real Time Semantic Segmentation On Cityscapes

評価指標

Frame (fps)
mIoU

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名
Frame (fps)
mIoU
Paper TitleRepository
STDC2-7597.0(1080Ti)76.8%Rethinking BiSeNet For Real-time Semantic Segmentation
Template-Based-NAS-arch01967.7%Template-Based Automatic Search of Compact Semantic Segmentation Architectures
DeepLab0.2563.1%Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
EDANet108.7 (1080Ti)67.3Efficient Dense Modules of Asymmetric Convolution for Real-Time Semantic Segmentation
DDRNet-23-slim101.6(2080Ti)77.4%Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes
U-HarDNet-70--HarDNet: A Low Memory Traffic Network
FasterSeg163.971.5%--
ICNet30.370.6%ICNet for Real-Time Semantic Segmentation on High-Resolution Images
PP-LiteSeg-B1195.3(1080Ti)73.9%PP-LiteSeg: A Superior Real-Time Semantic Segmentation Model
PP-LiteSeg-B2102.6(1080Ti)77.5%PP-LiteSeg: A Superior Real-Time Semantic Segmentation Model
ShelfNet1859.274.8%ShelfNet for Fast Semantic Segmentation
PIDNet-L31.1(3090)80.6%PIDNet: A Real-time Semantic Segmentation Network Inspired by PID Controllers
ESNet6370.7%ESNet: An Efficient Symmetric Network for Real-time Semantic Segmentation
ENet76.958.3%ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation
ENet + Lovász-Softmax76.963.1%The Lovász-Softmax loss: A tractable surrogate for the optimization of the intersection-over-union measure in neural networks
TD4-BISE1847.6 (Titan X)74.9%Temporally Distributed Networks for Fast Video Semantic Segmentation
BiSeNet V215672.6%BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation
RegSeg (no ImageNet pretraining)3078.3%Rethinking Dilated Convolution for Real-time Semantic Segmentation
PP-LiteSeg-T1273.6(1080Ti)72.0%PP-LiteSeg: A Superior Real-Time Semantic Segmentation Model
HyperSeg-M36.975.8%HyperSeg: Patch-wise Hypernetwork for Real-time Semantic Segmentation
0 of 39 row(s) selected.