DeepLab2: A TensorFlow Library for Deep Labeling

TL;DR

DeepLab2 is a comprehensive TensorFlow library for dense pixel prediction tasks in computer vision, providing state-of-the-art models, pretrained checkpoints, and evaluation tools to facilitate research and development.

Contribution

It introduces a unified TensorFlow library with multiple DeepLab model variants and pretrained models, enabling easier reproduction and advancement in dense pixel labeling research.

Findings

Achieves 68.0% PQ on Cityscapes validation set

Includes models with ImageNet-1K pretrained checkpoints

Facilitates future research with open-source code

Abstract

DeepLab2 is a TensorFlow library for deep labeling, aiming to provide a state-of-the-art and easy-to-use TensorFlow codebase for general dense pixel prediction problems in computer vision. DeepLab2 includes all our recently developed DeepLab model variants with pretrained checkpoints as well as model training and evaluation code, allowing the community to reproduce and further improve upon the state-of-art systems. To showcase the effectiveness of DeepLab2, our Panoptic-DeepLab employing Axial-SWideRNet as network backbone achieves 68.0% PQ or 83.5% mIoU on Cityscaspes validation set, with only single-scale inference and ImageNet-1K pretrained checkpoints. We hope that publicly sharing our library could facilitate future research on dense pixel labeling tasks and envision new applications of this technology. Code is made publicly available at \url{https://github.com/google-research/deeplab2}.