paper-with-me

Papers

Self-slimmed Vision Transformer

2021-11-24 · Zhuofan Zong, Kunchang Li, Guanglu Song, Yali Wang, Yu Qiao, Biao Leng, Yu Liu

Vision transformers (ViTs) have become the popular structures and outperformed convolutional neural networks (CNNs) on various vision tasks. However, such powerful transformers bring a huge computation burden, because of the exhausting token-to-token comparison. The previous works focus on dropping insignificant tokens to reduce the computational cost of ViTs. But when the dropping ratio increases, this hard manner will inevitably discard the vital tokens, which limits its efficiency. To solve the issue, we propose a generic self-slimmed learning approach for vanilla ViTs, namely SiT. Specifically, we first design a novel Token Slimming Module (TSM), which can boost the inference efficiency of ViTs by dynamic token aggregation. As a general method of token hard dropping, our TSM softly integrates redundant tokens into fewer informative ones. It can dynamically zoom visual attention without cutting off discriminative token relations in the images, even with a high slimming ratio. Furthermore, we introduce a concise Feature Recalibration Distillation (FRD) framework, wherein we design a reverse version of TSM (RTSM) to recalibrate the unstructured token in a flexible auto-encoder manner. Due to the similar structure between teacher and student, our FRD can effectively leverage structure knowledge for better convergence. Finally, we conduct extensive experiments to evaluate our SiT. It demonstrates that our method can speed up ViTs by 1.7x with negligible accuracy drop, and even speed up ViTs by 3.6x while maintaining 97% of their performance. Surprisingly, by simply arming LV-ViT with our SiT, we achieve new state-of-the-art performance on ImageNet. Code is available at https://github.com/Sense-X/SiT.

📄 PDF Abstract BibTeX arXiv:2111.12624

Code (1)

sense-x/sit 공식 구현 pytorch

Tasks

Knowledge Distillation

Methods 이 논문이 사용한 방법론

SPEED The monocular depth estimation (MDE) is the task of estimating depth from a single frame. This information is an essential knowledge in many computer vision tasks such as scene…
Knowledge Distillation A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions.…
LV-ViT 설명 없음

Similar Papers 제목 키워드 기반

Towards Efficient Adversarial Training on Vision Transformers

2022-07-21 · Boxi Wu, Jindong Gu, Zhifeng Li, Deng Cai 외

Vision Transformer (ViT), as a powerful alternative to Convolutional Neural Network (CNN), has received much attention. Recent work showed that ViTs are also vulnerable to adversarial examples like CNNs. To build robust …

Bilaterally Slimmable Transformer for Elastic and Efficient Visual Question Answering

2022-03-24 · Zhou Yu, Zitian Jin, Jun Yu, Mingliang Xu 외

Recent advances in Transformer architectures [1] have brought remarkable improvements to visual question answering (VQA). Nevertheless, Transformer-based VQA models are usually deep and wide to guarantee good performance…

GPUQuestion AnsweringVisual Question AnsweringVisual Question Answering (VQA)

LowFormer: Hardware Efficient Design for Convolutional Transformer Backbones

2024-09-05 · Moritz Nottebaum, Matteo Dunnhofer, Christian Micheloni

Research in efficient vision backbones is evolving into models that are a mixture of convolutions and transformer blocks. A smart combination of both, architecture-wise and component-wise is mandatory to excel in the spe…

CPUGPUImage Classificationobject-detection+2

DSPNet: Towards Slimmable Pretrained Networks based on Discriminative Self-supervised Learning

2022-07-13 · Shaoru Wang, Zeming Li, Jin Gao, Liang Li 외

Self-supervised learning (SSL) has achieved promising downstream performance. However, when facing various resource budgets in real-world applications, it costs a huge computation burden to pretrain multiple networks of …

Knowledge DistillationLinear evaluationSelf-Supervised Learning

SiT: Self-supervised vIsion Transformer

2021-04-08 · Sara Atito, Muhammad Awais, Josef Kittler

Self-supervised learning methods are gaining increasing traction in computer vision due to their recent success in reducing the gap with supervised learning. In natural language processing (NLP) self-supervised learning …

Few-Shot LearningSelf-Supervised Learning