ROI-based Deep Image Compression with Swin Transformers

by   Binglin Li, et al.
Xi'an Jiaotong University
Simon Fraser University

Encoding the Region Of Interest (ROI) with better quality than the background has many applications including video conferencing systems, video surveillance and object-oriented vision tasks. In this paper, we propose a ROI-based image compression framework with Swin transformers as main building blocks for the autoencoder network. The binary ROI mask is integrated into different layers of the network to provide spatial information guidance. Based on the ROI mask, we can control the relative importance of the ROI and non-ROI by modifying the corresponding Lagrange multiplier λ for different regions. Experimental results show our model achieves higher ROI PSNR than other methods and modest average PSNR for human evaluation. When tested on models pre-trained with original images, it has superior object detection and instance segmentation performance on the COCO validation dataset.


Vision Transformers Are Good Mask Auto-Labelers

We propose Mask Auto-Labeler (MAL), a high-quality Transformer-based mas...

Artistic Instance-Aware Image Filtering by Convolutional Neural Networks

In the recent years, public use of artistic effects for editing and beau...

Bottleneck Transformers for Visual Recognition

We present BoTNet, a conceptually simple yet powerful backbone architect...

Co-Scale Conv-Attentional Image Transformers

In this paper, we present Co-scale conv-attentional image Transformers (...

A Novel Algorithm for Exact Concave Hull Extraction

Region extraction is necessary in a wide range of applications, from obj...

A novel Region of Interest Extraction Layer for Instance Segmentation

Given the wide diffusion of deep neural network architectures for comput...

Accelerating Object Detection by Erasing Background Activations

Recent advances in deep learning have enabled complex real-world use cas...

Please sign up or login with your details

Forgot password? Click here to reset