Vision Transformer

Vision Transformers (ViTs) adapt the transformer architecture, initially designed for natural language processing, to image analysis by treating images as sequences of patches. Current research focuses on improving ViT efficiency and robustness through techniques like token pruning, attention engineering, and hybrid models combining ViTs with convolutional neural networks or other architectures (e.g., Mamba). These advancements are driving progress in various applications, including medical image analysis, object detection, and spatiotemporal prediction, by offering improved accuracy and efficiency compared to traditional convolutional neural networks in specific tasks.

Papers

November 6, 2024

November 4, 2024

November 1, 2024

Dual Low-Rank Adaptation for Continual Learning with Pre-Trained Models
Huancheng Chen, Jingtao Li, Nidham Gazagnadou, Weiming Zhuang, Chen Chen, Lingjuan Lyu
Vision Transformer Continual LEArning Pre Trained Model Continual Learning Adaptation Concern Low Rank Adaptation Dual System

October 30, 2024

October 29, 2024

October 28, 2024

Interpretable Image Classification with Adaptive Prototype-based Vision Transformers
Chiyu Ma, Jon Donnelly, Wenjun Liu, Soroush Vosoughi, Cynthia Rudin, Chaofan Chen
Vision Transformer Transformer Megatron Decepticons Deep Convolutional Neural Network Adaptive Importance Latent Prototype Interpretable Image Prototype Based Prototypical Part

October 27, 2024

October 26, 2024

Transforming Precision: A Comparative Analysis of Vision Transformers, CNNs, and Traditional ML for Knee Osteoarthritis Severity Diagnosis
Tasnim Sakib Apon, Md.Fahim-Ul-Islam, Nafiz Imtiaz Rafin, Joya Akter, Md. Golam Rabiul Alam
Vision Transformer CNN Model Image Processing Multidimensional Local Precision Rate Traditional Machine Learning ViT Model Knee Osteoarthritis Knee Osteoarthritis Severity

October 25, 2024

Multi-Class Abnormality Classification Task in Video Capsule Endoscopy
Dev Rishi Verma, Vibhor Saxena, Dhruv Sharma, Arpan Gupta
Vision Transformer Dual Attention Multi Class Anomaly Detection Capsule Endoscopy Multiscale Vision Transformer

Vision Transformer

Papers

Reducing catastrophic forgetting of incremental learning in the absence of rehearsal memory with task-specific token

Prion-ViT: Prions-Inspired Vision Transformers for Temperature prediction with Specklegrams

V-CAS: A Realtime Vehicle Anti Collision System Using Vision Transformer on Multi-Camera Streams

Learning Where to Edit Vision Transformers

Dual Low-Rank Adaptation for Continual Learning with Pre-Trained Models

ResiDual Transformer Alignment with Spectral Decomposition

ViT-LCA: A Neuromorphic Approach for Vision Transformers

Context-Aware Token Selection and Packing for Enhanced Vision Transformer

NMformer: A Transformer for Noisy Modulation Classification in Wireless Communication

DiaMond: Dementia Diagnosis with Multi-Modal Vision Transformers Using MRI and PET

Efficient Adaptation of Pre-trained Vision Transformer via Householder Transformation

FilterViT and DropoutViT: Lightweight Vision Transformer Models for Efficient Attention Mechanisms

Backdoor Attack Against Vision Transformers via Attention Gradient-Based Image Erosion

MAPUNetR: A Hybrid Vision Transformer and U-Net Architecture for Efficient and Interpretable Medical Image Segmentation

FakeFormer: Efficient Vulnerability-Driven Transformers for Generalisable Deepfake Detection

Interpretable Image Classification with Adaptive Prototype-based Vision Transformers

PViT: Prior-augmented Vision Transformer for Out-of-distribution Detection

Accelerating Augmentation Invariance Pretraining

Transforming Precision: A Comparative Analysis of Vision Transformers, CNNs, and Traditional ML for Knee Osteoarthritis Severity Diagnosis

Multi-Class Abnormality Classification Task in Video Capsule Endoscopy