# Analysis of Latent-Space Motion for Collaborative Intelligence

arXiv: https://arxiv.org/abs/2102.04018 · HTML: https://mlsanity.com/paper/2102.04018/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 22 references that are in the map.
- First posted on arXiv 8 Feb 2021.

## Builds on (most cited first, 12 of 12)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [JointDNN: An Efficient Training and Inference Engine for Intelligent Mobile Cloud Computing Services](https://mlsanity.com/paper/1801.08618/) (2018)
- [Video Coding for Machines: A Paradigm of Collaborative Compression and Intelligent Analytics](https://mlsanity.com/paper/2001.03569/) (2020)
- [Deep feature compression for collaborative object detection](https://mlsanity.com/paper/1802.03931/) (2018)
- [Multi-task learning with compressible features for Collaborative Intelligence](https://mlsanity.com/paper/1902.05179/) (2019)
- [Near-Lossless Deep Feature Compression for Collaborative Intelligence](https://mlsanity.com/paper/1804.09963/) (2018)
- [Pareto-Optimal Bit Allocation for Collaborative Intelligence](https://mlsanity.com/paper/2009.12430/) (2020)
- [Back-and-Forth prediction for deep tensor compression](https://mlsanity.com/paper/2002.07036/) (2020)
- [Deep Frame Prediction for Video Coding](https://mlsanity.com/paper/1901.00062/) (2018)
- [Bit Allocation for Multi-Task Collaborative Intelligence](https://mlsanity.com/paper/2002.07048/) (2020)
- [Shared Mobile-Cloud Inference for Collaborative Intelligence](https://mlsanity.com/paper/2002.00157/) (2020)

## Abstract (arXiv)

When the input to a deep neural network (DNN) is a video signal, a sequence of feature tensors is produced at the intermediate layers of the model. If neighboring frames of the input video are related through motion, a natural question is, "what is the relationship between the corresponding feature tensors?" By analyzing the effect of common DNN operations on optical flow, we show that the motion present in each channel of a feature tensor is approximately equal to the scaled version of the input motion. The analysis is validated through experiments utilizing common motion models. %These results will be useful in collaborative intelligence applications where sequences of feature tensors need to be compressed or further analyzed.