Fine-tuning visual models has been widely shown promising performance on many
downstream visual tasks. With the surprising development of pre-trained visual foundation …

Abstract Weakly-Supervised Semantic Segmentation (WSSS) using image-level labels
typically utilizes Class Activation Map (CAM) to generate the pseudo labels. Limited by the …

Transformer models have shown great success handling long-range interactions, making
them a promising tool for modeling video. However, they lack inductive biases and scale …

Abstract Sparsely activated Mixture-of-Experts (MoE) is becoming a promising paradigm for
multi-task learning (MTL). Instead of compressing multiple tasks' knowledge into a single …

DeepFake detection aims to differentiate falsified faces from real ones. Most approaches
formulate it as a binary classification problem by solely mining the local artifacts and …

Recently, the vision transformer and its variants have played an increasingly important role
in both monocular and multi-view human pose estimation. Considering image patches as …

The filtrations on Zhu's algebra $ A (V) $ and bimodules $ A (M) $ are studied. As an
application, we prove that $ A (V) $ is noetherian when $ V $ is strongly finitely generated …

Despite their remarkable achievement, gigantic transformers encounter significant
drawbacks, including exorbitant computational and memory footprints during training, as …

Abstract Recently efficient Vision Transformers have shown great performance with low
latency on resource-constrained devices. Conventionally they use 4x4 patch embeddings …

Vision transformers have recently set off a new wave in the field of medical image analysis
due to their remarkable performance on various computer vision tasks. However, recent …