The JPEG still picture compression standard

By Gregory K. Wallace. 1991

Subfields: Computer Vision and Pattern Recognition, Efficiency & Compression

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Efficiency & Compression

Citations in the corpus

Generating Diverse High-Fidelity Images with VQ-VAE-2 (2019, methodology): “Indeed, techniques such as JPEG [43] have shown that it is often possible to remove more than 80% of the data without noticeably changing the perceived image quality.”
On Buggy Resizing Libraries and Surprising Subtleties in FID Calculation (2021, methodology): “While the JPEG protocol is a lossy compression scheme, designed to preserve perceptual similarity to the original [73], it can perturb an image enough to corrupt downstream feature extraction.”
VideoGPT: Video Generation using VQ-VAE and Transformers (2021, neutral): “pixels across space and time? Below, we present our reasons for choosing the former: Natural images and videos contain a lot of spatial and temporal redundancies and hence the reason we use image compression tools such as JPEG (Wallace, 1992) and video codecs such as MPEG (Le Gall, 1991) everyday.”
Generating Images with Sparse Representations (2021, methodology): “In the JPEG codec, DCT blocks are quantized by dividing elementwise by a quantization matrix, and then rounding to the nearest integer.”
A practical tutorial on autoencoders for nonlinear feature fusion: Taxonomy, models, software and guidelines (2018, background): “This is the usual scenario while working with images, hence the popularity of the JPEG [120] graphic ﬁle format.”
Memory Replay with Data Compression for Continual Learning (2022, positive): “Here we evaluate the proposed idea by compressing images with JPEG (Wallace, 1992), a naive but commonly-used lossy compression algorithm.”
High-Fidelity Generative Image Compression (2020, methodology): “Various algorithms have been proposed over the years [50, 43, 54], including using stateof-the-art video compression algorithms for single image compression (BPG [7]).”
Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representations (2017, background): “SHA performs best on ImageNET100 and is most challenged on Kodak when compared with JPEG 2000.”
Compressing Convolutional Neural Networks in the Frequency Domain (2015, methodology): “Methods built on the DCT are widely used for compressing images and movies, including forming the standard technique for JPEG [29].”
Lossy Image Compression with Compressive Autoencoders (2017, methodology): “We compared our method to JPEG (Wallace, 1991), JPEG 2000 (Skodras et al., 2001), and the RNN-based method of (Toderici et al., 2016b) 4 .”
Real-Time Adaptive Image Compression (2017, neutral): “In traditional codecs such as JPEG and JP2, this is achieved via a pipeline which roughly breaks down into 3 modules: transformation, quantization, and encoding (Wallace (1992) and Rabbani & Joshi (2002) provide great overviews of the JPEG standards).”
On Aliased Resizing and Surprising Subtleties in GAN Evaluation (2022, methodology): “While the JPEG protocol is a lossy compression scheme, designed to preserve perceptual similarity to the original [67], it can perturb an image enough to corrupt downstream feature extraction.”
Learned Image Compression with Mixed Transformer-CNN Architectures (2023, background): “In the passing decades, there were many classical standards, including JPEG [31], WebP [11], and VVC [29], which contain three steps: transform, quantization, and entropy coding, have achieved impressive Rate-Distortion (RD) performance.”
End-to-End Learnt Image Compression via Non-Local Attention Optimization and Improved Context Modeling (2019, unknown): “In practice, some bit rate points cannot be precisely reached for BPG and JPEG.”
Variable Rate Deep Image Compression With a Conditional Autoencoder (2019, methodology): “Our experimental results show that the proposed scheme provides a better rate-distortion trade-off than the traditional variable-rate image compression codecs such as JPEG2000 and BPG.”
IMPACT: IMPrecise adders for low-power approximate computing (2011, methodology): “The Discrete Cosine Transform (DCT) and Inverse Discrete Cosine Transform (IDCT) are integral components of a JPEG image compression system [15].”
Learned Image Compression With Discretized Gaussian Mixture Likelihoods and Attention Modules (2020, methodology): “Classical image compression standards include JPEG [1], JPEG2000 [2], HEVC/H.265 [3] and ongoing Versatile Video Coding (VVC) [4] which will be the next generation compression standard expected by the end of 2020.”