Vision Transformer

2022-11-19 更新

CabViT: Cross Attention among Blocks for Vision Transformer

Authors:Haokui Zhang, Wenze Hu, Xiaoyu Wang

Since the vision transformer (ViT) has achieved impressive performance in image classification, an increasing number of researchers pay their attentions to designing more efficient vision transformer models. A general research line is reducing computational cost of self attention modules by adopting sparse attention or using local attention windows. In contrast, we propose to design high performance transformer based architectures by densifying the attention pattern. Specifically, we propose cross attention among blocks of ViT (CabViT), which uses tokens from previous blocks in the same stage as extra input to the multi-head attention of transformers. The proposed CabViT enhances the interactions of tokens across blocks with potentially different semantics, and encourages more information flows to the lower levels, which together improves model performance and model convergence with limited extra cost. Based on the proposed CabViT, we design a series of CabViT models which achieve the best trade-off between model size, computational cost and accuracy. For instance without the need of knowledge distillation to strength the training, CabViT achieves 83.0% top-1 accuracy on Imagenet with only 16.3 million parameters and about 3.9G FLOPs, saving almost half parameters and 13% computational cost while gaining 0.9% higher accuracy compared with ConvNext, use 52% of parameters but gaining 0.6% accuracy compared with distilled EfficientFormer
PDF 10 pages, 8 figures. Source code is available at


CAE v2: Context Autoencoder with CLIP Target

Authors:Xinyu Zhang, Jiahui Chen, Junkun Yuan, Qiang Chen, Jian Wang, Xiaodi Wang, Shumin Han, Xiaokang Chen, Jimin Pi, Kun Yao, Junyu Han, Errui Ding, Jingdong Wang

Masked image modeling (MIM) learns visual representation by masking and reconstructing image patches. Applying the reconstruction supervision on the CLIP representation has been proven effective for MIM. However, it is still under-explored how CLIP supervision in MIM influences performance. To investigate strategies for refining the CLIP-targeted MIM, we study two critical elements in MIM, i.e., the supervision position and the mask ratio, and reveal two interesting perspectives, relying on our developed simple pipeline, context autodecoder with CLIP target (CAE v2). Firstly, we observe that the supervision on visible patches achieves remarkable performance, even better than that on masked patches, where the latter is the standard format in the existing MIM methods. Secondly, the optimal mask ratio positively correlates to the model size. That is to say, the smaller the model, the lower the mask ratio needs to be. Driven by these two discoveries, our simple and concise approach CAE v2 achieves superior performance on a series of downstream tasks. For example, a vanilla ViT-Large model achieves 81.7% and 86.7% top-1 accuracy on linear probing and fine-tuning on ImageNet-1K, and 55.9% mIoU on semantic segmentation on ADE20K with the pre-training for 300 epochs. We hope our findings can be helpful guidelines for the pre-training in the MIM area, especially for the small-scale models.


CapEnrich: Enriching Caption Semantics for Web Images via Cross-modal Pre-trained Knowledge

Authors:Linli Yao, Weijing Chen, Qin Jin

Automatically generating textual descriptions for massive unlabeled images on the web can greatly benefit realistic web applications, e.g. multimodal retrieval and recommendation. However, existing models suffer from the problem of generating ``over-generic’’ descriptions, such as their tendency to generate repetitive sentences with common concepts for different images. These generic descriptions fail to provide sufficient textual semantics for ever-changing web images. Inspired by the recent success of Vision-Language Pre-training (VLP) models that learn diverse image-text concept alignment during pretraining, we explore leveraging their cross-modal pre-trained knowledge to automatically enrich the textual semantics of image descriptions. With no need for additional human annotations, we propose a plug-and-play framework, i.e CapEnrich, to complement the generic image descriptions with more semantic details. Specifically, we first propose an automatic data-building strategy to get desired training sentences, based on which we then adopt prompting strategies, i.e. learnable and template prompts, to incentivize VLP models to generate more textual details. For learnable templates, we fix the whole VLP model and only tune the prompt vectors, which leads to two advantages: 1) the pre-training knowledge of VLP models can be reserved as much as possible to describe diverse visual concepts; 2) only lightweight trainable parameters are required, so it is friendly to low data resources. Extensive experiments show that our method significantly improves the descriptiveness and diversity of generated sentences for web images. Our code will be released.
PDF Under Review


MultiCrossViT: Multimodal Vision Transformer for Schizophrenia Prediction using Structural MRI and Functional Network Connectivity Data

Authors:Yuda Bi, Anees Abrol, Zening Fu, Vince Calhoun

Vision Transformer (ViT) is a pioneering deep learning framework that can address real-world computer vision issues, such as image classification and object recognition. Importantly, ViTs are proven to outperform traditional deep learning models, such as convolutional neural networks (CNNs). Relatively recently, a number of ViT mutations have been transplanted into the field of medical imaging, thereby resolving a variety of critical classification and segmentation challenges, especially in terms of brain imaging data. In this work, we provide a novel multimodal deep learning pipeline, MultiCrossViT, which is capable of analyzing both structural MRI (sMRI) and static functional network connectivity (sFNC) data for the prediction of schizophrenia disease. On a dataset with minimal training subjects, our novel model can achieve an AUC of 0.832. Finally, we visualize multiple brain regions and covariance patterns most relevant to schizophrenia based on the resulting ViT attention maps by extracting features from transformer encoders.


Cross-domain Federated Adaptive Prompt Tuning for CLIP

Authors:Shangchao Su, Mingzhao Yang, Bin Li, Xiangyang Xue

Federated learning (FL) allows multiple parties to collaboratively train a global model without disclosing their data. Existing research often requires all model parameters to participate in the training procedure. However, with the advent of powerful pre-trained models, it becomes possible to achieve higher performance with fewer learnable parameters in FL. In this paper, we propose a federated adaptive prompt tuning algorithm, FedAPT, for cross-domain federated image classification scenarios with the vision-language pre-trained model, CLIP, which gives play to the strong representation ability in FL. Compared with direct federated prompt tuning, our core idea is to adaptively unlock specific domain knowledge for each test sample in order to provide them with personalized prompts. To implement this idea, we design an adaptive prompt tuning module, which consists of a global prompt, an adaptive network, and some keys. The server randomly generates a set of keys and assigns a unique key to each client. Then all clients cooperatively train the global adaptive network and global prompt with the local datasets and the frozen keys. Ultimately, the global aggregation model can assign a personalized prompt to CLIP based on the domain features of each test sample. We perform extensive experiments on two multi-domain image classification datasets. The results show that FedAPT can achieve better performance with less than 10\% of the number of parameters of the fully trained model, and the global model can perform well in different client domains simultaneously.


Demystify Self-Attention in Vision Transformers from a Semantic Perspective: Analysis and Application

Authors:Leijie Wu, Song Guo, Yaohong Ding, Junxiao Wang, Wenchao Xu, Richard Yida Xu, Jie Zhang

Self-attention mechanisms, especially multi-head self-attention (MSA), have achieved great success in many fields such as computer vision and natural language processing. However, many existing vision transformer (ViT) works simply inherent transformer designs from NLP to adapt vision tasks, while ignoring the fundamental difference between how MSA works in image and language settings''. Language naturally contains highly semantic structures that are directly interpretable by humans. Its basic unit (word) is discrete without redundant information, which readily supports interpretable studies on MSA mechanisms of language transformer. In contrast, visual data exhibits a fundamentally different structure: Its basic unit (pixel) is a natural low-level representation with significant redundancies in the neighbourhood, which poses obvious challenges to the interpretability of MSA mechanism in ViT. In this paper, we introduce a typical image processing technique, i.e., scale-invariant feature transforms (SIFTs), which maps low-level representations into mid-level spaces, and annotates extensive discrete keypoints with semantically rich information. Next, we construct a weighted patch interrelation analysis based on SIFT keypoints to capture the attention patterns hidden in patches with different semantic concentrations Interestingly, we find this quantitative analysis is not only an effective complement to the interpretability of MSA mechanisms in ViT, but can also be applied to 1) spurious correlation discovery andprompting’’ during model inference, 2) and guided model pre-training acceleration. Experimental results on both applications show significant advantages over baselines, demonstrating the efficacy of our method.
PDF 10 pages, 11 figures


All are Worth Words: A ViT Backbone for Diffusion Models

Authors:Fan Bao, Shen Nie, Kaiwen Xue, Yue Cao, Chongxuan Li, Hang Su, Jun Zhu

Vision transformers (ViT) have shown promise in various vision tasks while the U-Net based on a convolutional neural network (CNN) remains dominant in diffusion models. We design a simple and general ViT-based architecture (named U-ViT) for image generation with diffusion models. U-ViT is characterized by treating all inputs including the time, condition and noisy image patches as tokens and employing long skip connections between shallow and deep layers. We evaluate U-ViT in unconditional and class-conditional image generation, as well as text-to-image generation tasks, where U-ViT is comparable if not superior to a CNN-based U-Net of a similar size. In particular, a latent diffusion model with a small U-ViT achieves a record-breaking FID of 5.48 in text-to-image generation on MS-COCO, among methods without accessing large external datasets during the training of generative models. Besides, our results suggest that, for diffusion-based image modeling, the long skip connection is crucial while the down-sampling and up-sampling operators in CNN-based U-Net are not always necessary. We believe that U-ViT can provide insights for future research on backbones in diffusion models and benefit generative modeling on large scale cross-modality datasets.


What Images are More Memorable to Machines?

Authors:Junlin Han, Huangying Zhan, Jie Hong, Pengfei Fang, Hongdong Li, Lars Petersson, Ian Reid

This paper studies the problem of measuring and predicting how memorable an image is to pattern recognition machines, as a path to explore machine intelligence. Firstly, we propose a self-supervised machine memory quantification pipeline, dubbed ``MachineMem measurer’’, to collect machine memorability scores of images. Similar to humans, machines also tend to memorize certain kinds of images, whereas the types of images that machines and humans memorialize are different. Through in-depth analysis and comprehensive visualizations, we gradually unveil that “complex” images are usually more memorable to machines. We further conduct extensive experiments across 11 different machines (from linear classifiers to modern ViTs) and 9 pre-training methods to analyze and understand machine memory. This work proposes the concept of machine memorability and opens a new research direction at the interface between machine memory and visual data.
PDF Code: Project page:


CPT-V: A Contrastive Approach to Post-Training Quantization of Vision Transformers

Authors:Natalia Frumkin, Dibakar Gope, Diana Marculescu

When considering post-training quantization, prior work has typically focused on developing a mixed precision scheme or learning the best way to partition a network for quantization. In our work, CPT-V, we look at a general way to improve the accuracy of networks that have already been quantized, simply by perturbing the quantization scales. Borrowing the idea of contrastive loss from self-supervised learning, we find a robust way to jointly minimize a loss function using just 1,000 calibration images. In order to determine the best performing quantization scale, CPT-V contrasts the features of quantized and full precision models in a self-supervised fashion. Unlike traditional reconstruction-based loss functions, the use of a contrastive loss function not only rewards similarity between the quantized and full precision outputs but also helps in distinguishing the quantized output from other outputs within a given batch. In addition, in contrast to prior works, CPT-V proposes a block-wise evolutionary search to minimize a global contrastive loss objective, allowing for accuracy improvement of existing vision transformer (ViT) quantization schemes. For example, CPT-V improves the top-1 accuracy of a fully quantized ViT-Base by 10.30%, 0.78%, and 0.15% for 3-bit, 4-bit, and 8-bit weight quantization levels. Extensive experiments on a variety of other ViT architectures further demonstrate its robustness in extreme quantization scenarios. Our code is available at .


End-to-End Machine Learning Framework for Facial AU Detection in Intensive Care Units

Authors:Subhash Nerella, Kia Khezeli, Andrea Davidson, Patrick Tighe, Azra Bihorac, Parisa Rashidi

Pain is a common occurrence among patients admitted to Intensive Care Units. Pain assessment in ICU patients still remains a challenge for clinicians and ICU staff, specifically in cases of non-verbal sedated, mechanically ventilated, and intubated patients. Current manual observation-based pain assessment tools are limited by the frequency of pain observations administered and are subjective to the observer. Facial behavior is a major component in observation-based tools. Furthermore, previous literature shows the feasibility of painful facial expression detection using facial action units (AUs). However, these approaches are limited to controlled or semi-controlled environments and have never been validated in clinical settings. In this study, we present our Pain-ICU dataset, the largest dataset available targeting facial behavior analysis in the dynamic ICU environment. Our dataset comprises 76,388 patient facial image frames annotated with AUs obtained from 49 adult patients admitted to ICUs at the University of Florida Health Shands hospital. In this work, we evaluated two vision transformer models, namely ViT and SWIN, for AU detection on our Pain-ICU dataset and also external datasets. We developed a completely end-to-end AU detection pipeline with the objective of performing real-time AU detection in the ICU. The SWIN transformer Base variant achieved 0.88 F1-score and 0.85 accuracy on the held-out test partition of the Pain-ICU dataset.


文章作者: 木子已
版权声明: 本博客所有文章除特別声明外,均采用 CC BY 4.0 许可协议。转载请注明来源 木子已 !