Real-Time Semantic Segmentation Edit Task
Computer Vision • Semantic Segmentation
About
Real-time semantic segmentation is the task of achieving computationally efficient semantic segmentation (while maintaining a base level of accuracy).
( Image credit: TorchSeg )
Benchmarks
| TREND | DATASET | BEST METHOD | PAPER TITLE | PAPER | CODE | COMPARE |
|---|---|---|---|---|---|---|
|
||||||
|
||||||
|
||||||
|
||||||
|
Datasets
Greatest papers with code
Pyramid Scene Parsing Network
CVPR 2017
• tensorflow/models
• 
Scene parsing is challenging for unrestricted open vocabulary and diverse scenes.
Ranked #3 on
Video Semantic Segmentation
on Cityscapes val
LESION SEGMENTATION REAL-TIME SEMANTIC SEGMENTATION SCENE PARSING VIDEO SEMANTIC SEGMENTATION
Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
This is due to the very invariance properties that make DCNNs good for high level tasks.
Ranked #2 on
Scene Segmentation
on SUN-RGBD
Fully Convolutional Networks for Semantic Segmentation
Convolutional networks are powerful visual models that yield hierarchies of features.
Ranked #4 on
Scene Segmentation
on SUN-RGBD
REAL-TIME SEMANTIC SEGMENTATION SCENE SEGMENTATION VIDEO SEMANTIC SEGMENTATION
Feature Pyramid Encoding Network for Real-time Semantic Segmentation
• osmr/imgclsmob
• 
Although current deep learning methods have achieved impressive results for semantic segmentation, they incur high computational costs and have a huge number of parameters.
HarDNet: A Low Memory Traffic Network
We propose a Harmonic Densely Connected Network to achieve high efficiency in terms of both low MACs and memory traffic.
Ranked #2 on
Real-Time Semantic Segmentation
on Cityscapes test
DABNet: Depth-wise Asymmetric Bottleneck for Real-time Semantic Segmentation
• osmr/imgclsmob
•
As a pixel-level prediction task, semantic segmentation needs large computational cost with enormous parameters to obtain high performance.
ESNet: An Efficient Symmetric Network for Real-time Semantic Segmentation
• osmr/imgclsmob
•
The whole network has nearly symmetric architecture, which is mainly composed of a series of factorized convolution unit (FCU) and its parallel counterparts (PFCU).
Ranked #12 on
Real-Time Semantic Segmentation
on Cityscapes test
DiCENet: Dimension-wise Convolutions for Efficient Networks
• osmr/imgclsmob
•
When DiCE units are stacked to build the DiCENet model, we observe significant improvements over state-of-the-art models across various computer vision tasks including image classification, object detection, and semantic segmentation.
Ranked #17 on
Semantic Segmentation
on PASCAL VOC 2012 val
4 IMAGE CLASSIFICATION NEURAL ARCHITECTURE SEARCH REAL-TIME OBJECT DETECTION REAL-TIME SEMANTIC SEGMENTATION
LEDNet: A Lightweight Encoder-Decoder Network for Real-Time Semantic Segmentation
• osmr/imgclsmob
•
LEDNet: A Lightweight Encoder-Decoder Network for Real-time Semantic Segmentation
Ranked #13 on
Real-Time Semantic Segmentation
on Cityscapes test
Fast-SCNN: Fast Semantic Segmentation Network
• osmr/imgclsmob
•
The encoder-decoder framework is state-of-the-art for offline semantic image segmentation.
Ranked #27 on
Semantic Segmentation
on Cityscapes val