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

August 4, 2024

Human-inspired Explanations for Vision Transformers and Convolutional Neural Networks
Mahadev Prasad Panda, Matteo Tiezzi, Martina Vilas, Gemma Roig, Bjoern M. Eskofier, Dario Zanca
Convolutional Neural Network Vision Transformer Deep Convolutional Neural Network High Explainability Explanation Map Human Like Explanation

August 3, 2024

LAM3D: Leveraging Attention for Monocular 3D Object Detection
Diana-Alexandra Sas, Leandro Di Bella, Yangxintong Lyu, Florin Oniga, Adrian Munteanu
Vision Transformer Human Attention Monocular 3D Object Detection 3D Object Detection Benchmark Pyramid Vision Transformer

August 2, 2024

August 1, 2024

Leaf Angle Estimation using Mask R-CNN and LETR Vision Transformer
Venkat Margapuri, Prapti Thapaliya, Trevor Rife
Vision Transformer Mask R CNN Plant Growth Simulation High Throughput Phenotyping Leaf Image Leaf Size

July 31, 2024

July 29, 2024

July 28, 2024

July 27, 2024

Channel Boosted CNN-Transformer-based Multi-Level and Multi-Scale Nuclei Segmentation
Zunaira Rauf, Abdul Rehman Khan, Asifullah Khan
Vision Transformer Nucleus Segmentation Channel Importance

July 26, 2024

July 25, 2024

Vision Transformer

Papers

Human-inspired Explanations for Vision Transformers and Convolutional Neural Networks

LAM3D: Leveraging Attention for Monocular 3D Object Detection

Transformers are Universal In-context Learners

Privacy-Preserving Split Learning with Vision Transformers using Patch-Wise Random and Noisy CutMix

Leaf Angle Estimation using Mask R-CNN and LETR Vision Transformer

An Explainable Vision Transformer with Transfer Learning Combined with Support Vector Machine Based Efficient Drought Stress Identification

SimpleLLM4AD: An End-to-End Vision-Language Model with Graph Visual Question Answering for Autonomous Driving

MimiQ: Low-Bit Data-Free Quantization of Vision Transformers with Encouraging Inter-Head Attention Similarity

Mixture of Nested Experts: Adaptive Processing of Visual Tokens

Twins-PainViT: Towards a Modality-Agnostic Vision Transformer Framework for Multimodal Automatic Pain Assessment using Facial Videos and fNIRS

Rethinking RGB-D Fusion for Semantic Segmentation in Surgical Datasets

Combined CNN and ViT features off-the-shelf: Another astounding baseline for recognition

Depth-Wise Convolutions in Vision Transformers for Efficient Training on Small Datasets

Channel Boosted CNN-Transformer-based Multi-Level and Multi-Scale Nuclei Segmentation

VSSD: Vision Mamba with Non-Causal State Space Duality

Skin Cancer Detection utilizing Deep Learning: Classification of Skin Lesion Images using a Vision Transformer

Mixed Non-linear Quantization for Vision Transformers

Quasar-ViT: Hardware-Oriented Quantization-Aware Architecture Search for Vision Transformers

HG-PIPE: Vision Transformer Acceleration with Hybrid-Grained Pipeline

How Lightweight Can A Vision Transformer Be