DCT-CompCNN: A Novel Image Classification Network Using JPEG Compressed DCT Coefficients

TL;DR

This paper introduces DCT-CompCNN, a new CNN architecture that directly uses JPEG compressed DCT coefficients as input, demonstrating improved image classification performance over traditional RGB-based CNNs.

Contribution

The paper proposes a novel CNN model that processes JPEG DCT coefficients directly, enhancing classification accuracy without decompressing images.

Findings

DCT-CompCNN outperforms conventional CNNs on benchmark datasets.

Using JPEG compressed data as input can improve classification accuracy.

The approach is compatible with existing architectures like ResNet-50.

Abstract

The popularity of Convolutional Neural Network (CNN) in the field of Image Processing and Computer Vision has motivated researchers and industrialist experts across the globe to solve different challenges with high accuracy. The simplest way to train a CNN classifier is to directly feed the original RGB pixels images into the network. However, if we intend to classify images directly with its compressed data, the same approach may not work better, like in case of JPEG compressed images. This research paper investigates the issues of modifying the input representation of the JPEG compressed data, and then feeding into the CNN. The architecture is termed as DCT-CompCNN. This novel approach has shown that CNNs can also be trained with JPEG compressed DCT coefficients, and subsequently can produce a better performance in comparison with the conventional CNN approach. The efficiency of the modified input representation is tested with the existing ResNet-50 architecture and the proposed DCT-CompCNN architecture on a public image classification datasets like Dog Vs Cat and CIFAR-10 datasets, reporting a better performance