Efficient Video Semantic Segmentation with Labels Propagation and Refinement

TL;DR

This paper introduces EVS, a hybrid CPU/GPU pipeline for real-time high-definition video semantic segmentation that combines optical flow, CNNs, and a refinement module to achieve high accuracy and frame rates.

Contribution

The paper presents a novel hybrid approach integrating optical flow and CNNs with a refinement module for efficient real-time video segmentation.

Findings

Achieves over 60% mIoU on Cityscapes dataset.

Operates at 80 to 1000 Hz on a single GPU/CPU.

Outperforms existing real-time segmentation methods in speed.

Abstract

This paper tackles the problem of real-time semantic segmentation of high definition videos using a hybrid GPU / CPU approach. We propose an Efficient Video Segmentation(EVS) pipeline that combines: (i) On the CPU, a very fast optical flow method, that is used to exploit the temporal aspect of the video and propagate semantic information from one frame to the next. It runs in parallel with the GPU. (ii) On the GPU, two Convolutional Neural Networks: A main segmentation network that is used to predict dense semantic labels from scratch, and a Refiner that is designed to improve predictions from previous frames with the help of a fast Inconsistencies Attention Module (IAM). The latter can identify regions that cannot be propagated accurately. We suggest several operating points depending on the desired frame rate and accuracy. Our pipeline achieves accuracy levels competitive to the existing real-time methods for semantic image segmentation(mIoU above 60%), while achieving much higher frame rates. On the popular Cityscapes dataset with high resolution frames (2048 x 1024), the proposed operating points range from 80 to 1000 Hz on a single GPU and CPU.