A Comprehensive Study of Vision Transformers on Dense Prediction Tasks
원문 초록
Convolutional Neural Networks (CNNs), architectures consisting of convolutional layers, have been the standard choice in vision tasks. Recent studies have shown that Vision Transformers (VTs), architectures based on self-attention modules, achieve comparable performance in challenging tasks such as object detection and semantic segmentation. However, the image processing mechanism of VTs is different from that of conventional CNNs. This poses several questions about their generalizability, robustness, reliability, and texture bias when used to extract features for complex tasks. To address these questions, we study and compare VT and CNN architectures as feature extractors in object detection and semantic segmentation. Our extensive empirical results show that the features generated by VTs are more robust to distribution shifts, natural corruptions, and adversarial attacks in both tasks, whereas CNNs perform better at higher image resolutions in object detection. Furthermore, our results demonstrate that VTs in dense prediction tasks produce more reliable and less texture-biased predictions.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Automatic Traffic Sign Detection and Recognition Using SegU-Net and a Modified Tversky Loss Function With L1-Constraint
Beyond One Shot, Beyond One Perspective: Cross-View and Long-Horizon Distillation for Better LiDAR Representations
NOCTIS: Novel Object Cyclic Threshold based Instance Segmentation
Seg-R1: Segmentation Can Be Surprisingly Simple with Reinforcement Learning
SeC: Advancing Complex Video Object Segmentation via Progressive Concept Construction
A Real-Time System for Egocentric Hand-Object Interaction Detection in Industrial Domains
토론
이 논문과 연결된 포럼 글이 아직 없습니다.