Log In Sign Up

Pruned Lightweight Encoders for Computer Vision

by   Jakub Žádník, et al.

Latency-critical computer vision systems, such as autonomous driving or drone control, require fast image or video compression when offloading neural network inference to a remote computer. To ensure low latency on a near-sensor edge device, we propose the use of lightweight encoders with constant bitrate and pruned encoding configurations, namely, ASTC and JPEG XS. Pruning introduces significant distortion which we show can be recovered by retraining the neural network with compressed data after decompression. Such an approach does not modify the network architecture or require coding format modifications. By retraining with compressed datasets, we reduced the classification accuracy and segmentation mean intersection over union (mIoU) degradation due to ASTC compression to 4.9-5.0 percentage points (pp) and 4.4-4.0 pp, respectively. With the same method, the mIoU lost due to JPEG XS compression at the main profile was restored to 2.7-2.3 pp. In terms of encoding speed, our ASTC encoder implementation is 2.3x faster than JPEG. Even though the JPEG XS reference encoder requires optimizations to reach low latency, we showed that disabling significance flag coding saves 22-23 0.4-0.3 mIoU after retraining.


page 1

page 5


Task Oriented Video Coding: A Survey

Video coding technology has been continuously improved for higher compre...

Practical Network Acceleration with Tiny Sets

Network compression is effective in accelerating the inference of deep n...

Attention-based Feature Compression for CNN Inference Offloading in Edge Computing

This paper studies the computational offloading of CNN inference in devi...

Learned Image Compression for Machine Perception

Recent work has shown that learned image compression strategies can outp...

Variable-Byte Encoding is Now Space-Efficient Too

The ubiquitous Variable-Byte encoding is considered one of the fastest c...

WeightMom: Learning Sparse Networks using Iterative Momentum-based pruning

Deep Neural Networks have been used in a wide variety of applications wi...

Dynamic Encoder Transducer: A Flexible Solution For Trading Off Accuracy For Latency

We propose a dynamic encoder transducer (DET) for on-device speech recog...