# Efficient Feature Compression for Machines with Global Statistics Preservation

arXiv: https://arxiv.org/abs/2512.09235 · HTML: https://mlsanity.com/paper/2512.09235/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 20 references that are in the map.
- First posted on arXiv 10 Dec 2025.

## Builds on (most cited first, 9 of 9)

- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [YOLOv3: An Incremental Improvement](https://mlsanity.com/paper/1804.02767/) (2018)
- [Towards Real-Time Multi-Object Tracking](https://mlsanity.com/paper/1909.12605/) (2019)
- [Tensor Programs I: Wide Feedforward or Recurrent Neural Networks of Any Architecture are Gaussian Processes](https://mlsanity.com/paper/1910.12478/) (2019)
- [Large-scale interactive object segmentation with human annotators](https://mlsanity.com/paper/1903.10830/) (2019)
- [Collaborative Inference for AI-Empowered IoT Devices](https://mlsanity.com/paper/2207.11664/) (2022)
- [Lightweight Compression of Intermediate Neural Network Features for Collaborative Intelligence](https://mlsanity.com/paper/2105.07102/) (2021)

## Built on by (most cited first, 3 of 3)

- [Emerging Standards for Machine-to-Machine Video Coding](https://mlsanity.com/paper/2512.10230/) (2025)
- [Feature Coding for Scalable Machine Vision](https://mlsanity.com/paper/2512.10209/) (2025)
- [Towards Large Model Feature Coding](https://mlsanity.com/paper/2605.24025/) (2026)

## Abstract (arXiv)

The split-inference paradigm divides an artificial intelligence (AI) model into two parts. This necessitates the transfer of intermediate feature data between the two halves. Here, effective compression of the feature data becomes vital. In this paper, we employ Z-score normalization to efficiently recover the compressed feature data at the decoder side. To examine the efficacy of our method, the proposed method is integrated into the latest Feature Coding for Machines (FCM) codec standard under development by the Moving Picture Experts Group (MPEG). Our method supersedes the existing scaling method used by the current standard under development. It both reduces the overhead bits and improves the end-task accuracy. To further reduce the overhead in certain circumstances, we also propose a simplified method. Experiments show that using our proposed method shows 17.09% reduction in bitrate on average across different tasks and up to 65.69% for object tracking without sacrificing the task accuracy.