When are Foundation Models Effective? Understanding the Suitability for Pixel-Level Classification Using Multispectral Imageryby Yiqun…
Supervised Contrastive Vision Transformer for Breast Histopathological Image Classificationby Mohammad Shiri, Monalika Padma Reddy, Jiangwen…
Masked Autoencoders for Microscopy are Scalable Learners of Cellular Biologyby Oren Kraus, Kian Kenyon-Dean, Saber…
Rethinking Low-Rank Adaptation in Vision: Exploring Head-Level Responsiveness across Diverse Tasksby Yibo Zhong, Yao ZhouFirst…
Single-image driven 3d viewpoint training data augmentation for effective wine label recognitionby Yueh-Cheng Huang, Hsin-Yi…
A Multi-Level Framework for Accelerating Training Transformer Modelsby Longwei Zou, Han Zhang, Yangdong DengFirst submitted…
OmniFusion Technical Reportby Elizaveta Goncharova, Anton Razzhigaev, Matvey Mikhalchuk, Maxim Kurkin, Irina Abdullaeva, Matvey Skripkin,…
ActNetFormer: Transformer-ResNet Hybrid Method for Semi-Supervised Action Recognition in Videosby Sharana Dharshikgan Suresh Dass, Hrishav…
A Lightweight Measure of Classification Difficulty from Application Dataset Characteristicsby Bryan Bo Cao, Abhinav Sharma,…
DeiT-LT Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasetsby Harsh Rangwani, Pradipto Mondal,…