English
Related papers

Related papers: SoftVQ-VAE: Efficient 1-Dimensional Continuous Tok…

200 papers

Visual Autoregressive (VAR) modeling has gained popularity for its shift towards next-scale prediction. However, existing VAR paradigms process the entire token map at each scale step, leading to the complexity and runtime scaling…

Computer Vision and Pattern Recognition · Computer Science 2025-07-09 Hang Guo , Yawei Li , Taolin Zhang , Jiangshan Wang , Tao Dai , Shu-Tao Xia , Luca Benini

Deep neural networks with discrete latent variables offer the promise of better symbolic reasoning, and learning abstractions that are more useful to new tasks. There has been a surge in interest in discrete latent variable models, however,…

Machine Learning · Computer Science 2018-07-23 Aurko Roy , Ashish Vaswani , Arvind Neelakantan , Niki Parmar

Recent research has shown a strong theoretical connection between variational autoencoders (VAEs) and the rate-distortion theory. Motivated by this, we consider the problem of lossy image compression from the perspective of generative…

Image and Video Processing · Electrical Eng. & Systems 2023-03-28 Zhihao Duan , Ming Lu , Zhan Ma , Fengqing Zhu

Whilst diffusion probabilistic models can generate high quality image content, key limitations remain in terms of both generating high-resolution imagery and their associated high computational requirements. Recent Vector-Quantized image…

Computer Vision and Pattern Recognition · Computer Science 2021-11-25 Sam Bond-Taylor , Peter Hessey , Hiroshi Sasaki , Toby P. Breckon , Chris G. Willcocks

In this paper we introduce learnable lattice vector quantization and demonstrate its effectiveness for learning discrete representations. Our method, termed LL-VQ-VAE, replaces the vector quantization layer in VQ-VAE with lattice-based…

Machine Learning · Computer Science 2023-10-17 Ahmed Khalil , Robert Piechocki , Raul Santos-Rodriguez

Recent Multimodal Large Language Models (MLLMs) have demonstrated strong performance on vision-language understanding tasks, yet their inference efficiency is often hampered by the large number of visual tokens, particularly in…

Computer Vision and Pattern Recognition · Computer Science 2026-04-21 Jiafei Song , Fengwei Zhou , Jin Qu , Wenjin Jason Li , Tong Wu , Gengjian Xue , Zhikang Zhao , Daomin Wei , Yichao Lu , Bailin Na

Autoregressive (AR) video generative models rely on video tokenizers that compress pixels into discrete token sequences. The length of these token sequences is crucial for balancing reconstruction quality against downstream generation…

Computer Vision and Pattern Recognition · Computer Science 2026-03-13 Tianwei Xiong , Jun Hao Liew , Zilong Huang , Zhijie Lin , Jiashi Feng , Xihui Liu

VQ-based image generation typically follows a two-stage pipeline: a tokenizer encodes images into discrete tokens, and a generative model learns their dependencies for reconstruction. However, improved tokenization in the first stage does…

Computer Vision and Pattern Recognition · Computer Science 2026-02-02 Bin Wu , Mengqi Huang , Weinan Jia , Zhendong Mao

Image compression at extremely low bitrates (below 0.1 bits per pixel (bpp)) is a significant challenge due to substantial information loss. In this work, we propose a novel two-stage extreme image compression framework that exploits the…

Image and Video Processing · Electrical Eng. & Systems 2024-09-05 Zhiyuan Li , Yanhui Zhou , Hao Wei , Chenyang Ge , Jingwen Jiang

High-resolution Large Multimodal Models (LMMs) encounter the challenges of excessive visual tokens and quadratic visual complexity. Current high-resolution LMMs address the quadratic complexity while still generating excessive visual…

Computer Vision and Pattern Recognition · Computer Science 2024-05-27 Chunjiang Ge , Sijie Cheng , Ziming Wang , Jiale Yuan , Yuan Gao , Jun Song , Shiji Song , Gao Huang , Bo Zheng

Constructing powerful generative models for natural images is a challenging task. PixelCNN models capture details and local information in images very well but have limited receptive field. Variational autoencoders with a factorial decoder…

Computer Vision and Pattern Recognition · Computer Science 2019-08-28 Hossein Sadeghi , Evgeny Andriyash , Walter Vinci , Lorenzo Buffoni , Mohammad H. Amin

Image tokenization has significantly advanced visual generation and multimodal modeling, particularly when paired with autoregressive models. However, current methods face challenges in balancing efficiency and fidelity: high-resolution…

Computer Vision and Pattern Recognition · Computer Science 2025-03-17 Qingsong Xie , Zhao Zhang , Zhe Huang , Yanhao Zhang , Haonan Lu , Zhenyu Yang

Image compression under ultra-low bitrates remains challenging for both conventional learned image compression (LIC) and generative vector-quantized (VQ) modeling. Conventional LIC suffers from severe artifacts due to heavy quantization,…

Computer Vision and Pattern Recognition · Computer Science 2025-11-24 Lei Lu , Yize Li , Yanzhi Wang , Wei Wang , Wei Jiang

Variational Autoencoders (VAEs) are powerful generative models that have been widely used in various fields, including image and text generation. However, one of the known challenges in using VAEs is the model's sensitivity to its…

Machine Learning · Computer Science 2024-12-31 Gabriela Sejnova , Michal Vavrecka , Karla Stepanova

Latent diffusion models (LDMs) enable high-fidelity synthesis by operating in learned latent spaces. However, training state-of-the-art LDMs requires complex staging: a tokenizer must be trained first, before the diffusion model can be…

Computer Vision and Pattern Recognition · Computer Science 2026-03-24 Shivam Duggal , Xingjian Bai , Zongze Wu , Richard Zhang , Eli Shechtman , Antonio Torralba , Phillip Isola , William T. Freeman

Learning a robust video Variational Autoencoder (VAE) is essential for reducing video redundancy and facilitating efficient video generation. Directly applying image VAEs to individual frames in isolation can result in temporal…

Computer Vision and Pattern Recognition · Computer Science 2024-12-24 Yazhou Xing , Yang Fei , Yingqing He , Jingye Chen , Jiaxin Xie , Xiaowei Chi , Qifeng Chen

The high dimensionality of images presents architecture and sampling-efficiency challenges for likelihood-based generative models. Previous approaches such as VQ-VAE use deep autoencoders to obtain compact representations, which are more…

Computer Vision and Pattern Recognition · Computer Science 2021-03-08 Charlie Nash , Jacob Menick , Sander Dieleman , Peter W. Battaglia

Virtual humans have gained considerable attention in numerous industries, e.g., entertainment and e-commerce. As a core technology, synthesizing photorealistic face frames from target speech and facial identity has been actively studied…

Autoencoder (AE) is the key to the success of latent diffusion models for image and video generation, reducing the denoising resolution and improving efficiency. However, the power of AE has long been underexplored in terms of network…

Computer Vision and Pattern Recognition · Computer Science 2025-10-02 Yushu Wu , Yanyu Li , Ivan Skorokhodov , Anil Kag , Willi Menapace , Sharath Girish , Aliaksandr Siarohin , Yanzhi Wang , Sergey Tulyakov

Autoregressive and diffusion models drive the recent breakthroughs on text-to-image generation. Despite their huge success of generating high-realistic images, a common shortcoming of these models is their high inference latency -…

Computer Vision and Pattern Recognition · Computer Science 2023-12-27 Zhangyin Feng , Runyi Hu , Liangxin Liu , Fan Zhang , Duyu Tang , Yong Dai , Xiaocheng Feng , Jiwei Li , Bing Qin , Shuming Shi