# Semantic Segmentation in Learned Compressed Domain

arXiv: https://arxiv.org/abs/2209.01355 · HTML: https://mlsanity.com/paper/2209.01355/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 17 references that are in the map.
- First posted on arXiv 3 Sep 2022.

## Builds on (most cited first, 11 of 11)

- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [Rethinking Atrous Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1706.05587/) (2017)
- [Variational image compression with a scale hyperprior](https://mlsanity.com/paper/1802.01436/) (2018)
- [Image Segmentation Using Deep Learning: A Survey](https://mlsanity.com/paper/2001.05566/) (2020)
- [Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modules](https://mlsanity.com/paper/2001.01568/) (2020)
- [CompressAI: a PyTorch library and evaluation platform for end-to-end compression research](https://mlsanity.com/paper/2011.03029/) (2020)
- [Learning in the Frequency Domain](https://mlsanity.com/paper/2002.12416/) (2020)
- [Scalable Image Coding for Humans and Machines](https://mlsanity.com/paper/2107.08373/) (2021)
- [Towards Image Understanding from Deep Compression without Decoding](https://mlsanity.com/paper/1803.06131/) (2018)
- [Towards End-to-End Image Compression and Analysis with Transformers](https://mlsanity.com/paper/2112.09300/) (2021)

## Built on by (most cited first, 10 of 10)

- [Rate-Distortion-Cognition Controllable Versatile Neural Image Compression](https://mlsanity.com/paper/2407.11700/) (2024)
- [Image Coding for Machines with Object Region Learning](https://mlsanity.com/paper/2308.13984/) (2023)
- [Tell Codec What Worth Compressing: Semantically Disentangled Image Coding for Machine with LMMs](https://mlsanity.com/paper/2408.08575/) (2024)
- [Distributed Semantic Segmentation with Efficient Joint Source and Task Decoding](https://mlsanity.com/paper/2407.11224/) (2024)
- [Revisiting MLLM Token Technology through the Lens of Classical Visual Coding](https://mlsanity.com/paper/2508.13460/) (2025)
- [Compression Tells Intelligence: Visual Coding, Visual Token Technology, and the Unification](https://mlsanity.com/paper/2601.20742/) (2026)
- [An Efficient Semantic Segmentation Decoder for In-Car or Distributed Applications](https://mlsanity.com/paper/2510.16747/) (2025)
- [DNN-Compressed Domain Visual Recognition with Feature Adaptation](https://mlsanity.com/paper/2305.08000/) (2023)
- [CompTLL-UNet: Compressed Domain Text-Line Localization in Challenging Handwritten Documents using Deep Feature Learning from JPEG Coefficients](https://mlsanity.com/paper/2308.06142/) (2023)
- [Distributed Semantic Segmentation With Improved Rate-Distortion Trade-Off](https://mlsanity.com/paper/2608.28684/) (2026)

## Abstract (arXiv)

Most machine vision tasks (e.g., semantic segmentation) are based on images encoded and decoded by image compression algorithms (e.g., JPEG). However, these decoded images in the pixel domain introduce distortion, and they are optimized for human perception, making the performance of machine vision tasks suboptimal. In this paper, we propose a method based on the compressed domain to improve segmentation tasks. i) A dynamic and a static channel selection method are proposed to reduce the redundancy of compressed representations that are obtained by encoding. ii) Two different transform modules are explored and analyzed to help the compressed representation be transformed as the features in the segmentation network. The experimental results show that we can save up to 15.8% bitrates compared with a state-of-the-art compressed domain-based work while saving up to about 83.6% bitrates and 44.8% inference time compared with the pixel domain-based method.