Learning based Facial Image Compression with Semantic Fidelity Metric

  title={Learning based Facial Image Compression with Semantic Fidelity Metric},
  author={Zhibo Chen and Tianyu He},

Progressive Deep Image Compression for Hybrid Contexts of Image Classification and Reconstruction

The proposed DIC framework, called residual-enhanced mask-based progressive generative coding (RMPGC), is designed for explicit control of the performance within the rate-distortion-classification-perception (RDCP) trade-off, and is a flexible framework that can be applied to different neural network backbones.

A generative adversarial network for video compression

A novel inter-frame video coding scheme that compresses both reference frames and target (residue) frames by GAN and adopts adversarial learning, which effectively reduces the bit rates.

Towards an Efficient Facial Image Compression with Neural Networks

The results show that the proposed approach consists of a custom loss that combines the two tasks of image compression and face recognition guaranteeing high perceptive quality and face verification accuracy.

Multi-scale Grouped Dense Network for VVC Intra Coding

The multi-scale grouped dense network (MSGDN) is designed to further reduce the compression artifacts by combining the multi- scale and grouped dense block, which are integrated as the post-process network of VVC intra coding.

Cross Modal Compression: Towards Human-comprehensible Semantic Compression

The qualitative and quantitative results show that the proposed cross modal compression, a semantic compression framework for visual data, can achieve encouraging reconstructed results with an ultrahigh compression ratio, showing better compression performance than the widely used JPEG baseline.

Multi-rate deep semantic image compression with quantized modulated autoencoder

  • D. Sebai
  • Computer Science
    2021 IEEE 23rd International Workshop on Multimedia Signal Processing (MMSP)
  • 2021
An end-to-end multi-rate deep semantic image compression with quantized conditional autoencoder that includes two neural networks for the semantic analysis and image compression, respectively.

Learned Block-Based Hybrid Image Compression

This paper introduces explicit intra prediction into a learned image compression framework to utilize the relation among adjacent blocks and proposes a contextual prediction module (CPM) to better capture long-range correlations by utilizing the strip pooling to extract the most relevant information in neighboring latent space, thus achieving effective information prediction.

End-to-End Image Compression with Probabilistic Decoding

A learned image compression framework to natively support probabilistic decoding that is dependent on a revertible neural network-based transform to convert pixels into coefficients that obey the pre-chosen distribution as much as possible.

3-D Context Entropy Model for Improved Practical Image Compression

A 3-D context entropy model which can take advantage of known latent representation in current spatial locations for better entropy estimation is proposed and a light-weighted residual structure is adopted for feature learning during entropy estimation.

Task-Driven Semantic Coding via Reinforcement Learning

This work designs semantic maps for different tasks to extract the pixelwise semantic fidelity for videos/images and implements task-driven semantic coding by implementing semantic bit allocation based on reinforcement learning (RL).



Low Bit-Rate Compression of Facial Images

An efficient approach for face compression is introduced by restricting a family of images to frontal facial mug shots to compress facial images at very low bit rates while keeping high visual qualities, outperforming JPEG-2000 performance significantly.

Facial Image Compression using Patch-Ordering-Based Adaptive Wavelet Transform

This letter proposes a novel compression algorithm, exploiting the recently developed redundant tree-based wavelet transform, designed to best sparsify the whole set of aligned frontal facial images using a common feature-ordering.

Recognizable or Not: Towards Image Semantic Quality Assessment for Compression

This paper proposes a full-reference ISQA measure for image semantic quality assessment (ISQA), and performs subjective test about text recognition from compressed images, and observes that the measure has high consistency with subjective recognizability.

Improved Lossy Image Compression with Priming and Spatially Adaptive Bit Rates for Recurrent Networks

A method for lossy image compression based on recurrent, convolutional neural networks that outperforms BPG, WebP, JPEG2000, and JPEG as measured by MS-SSIM is proposed and it is shown that training with a pixel-wise loss weighted by SSIM increases reconstruction quality according to multiple metrics.

Image Compression in Face Recognition - a Literature Survey

It is shown that most researches agree that compression does not significantly deteriorate recognition accuracy in both cases, and a lot of work is still to be done to reach the real life implementation stage of compression in face recognition systems.

Full Resolution Image Compression with Recurrent Neural Networks

This is the first neural network architecture that is able to outperform JPEG at image compression across most bitrates on the rate-distortion curve on the Kodak dataset images, with and without the aid of entropy coding.

An End-to-End Compression Framework Based on Convolutional Neural Networks

Experimental results validate that the proposed compression framework greatly outperforms several compression frameworks that use existing image coding standards with the state-of-the-art deblocking or denoising post-processing methods.

Real-Time Adaptive Image Compression

A machine learning-based approach to lossy image compression which outperforms all existing codecs, while running in real-time, and supplementing the approach with adversarial training specialized towards use in a compression setting.

Variational image compression with a scale hyperprior

It is demonstrated that this model leads to state-of-the-art image compression when measuring visual quality using the popular MS-SSIM index, and yields rate-distortion performance surpassing published ANN-based methods when evaluated using a more traditional metric based on squared error (PSNR).

Conditional Probability Models for Deep Image Compression

This paper proposes a new technique to navigate the rate-distortion trade-off for an image compression auto-encoder by using a context model: A 3D-CNN which learns a conditional probability model of the latent distribution of the auto- Encoder.