Transformer-based architectures have become competitive across a variety of visual
domains, most notably images and videos. While prior work studies these modalities in …

With the exponential growth of video data, there is an urgent need for automated technology
to analyze and comprehend video content. However, existing video understanding models …

This paper proposes a self-supervised approach to learn universal facial representations
from videos, that can transfer across a variety of facial analysis tasks such as Facial Attribute …

We present a simple approach which can turn a ViT encoder into an efficient video model,
which can seamlessly work with both image and video inputs. By sparsely sampling the …

While transformers have begun to dominate many tasks in vision, applying them to large
images is still computationally difficult. A large reason for this is that self-attention scales …

Visual segmentation seeks to partition images, video frames, or point clouds into multiple
segments or groups. This technique has numerous real-world applications, such as …

There has been a longstanding belief that generation can facilitate a true understanding of
visual data. In line with this, we revisit generatively pre-training visual representations in light …

Addressing the dual challenges of local redundancy and global dependencies in video
understanding, this work innovatively adapts the Mamba to the video domain. The proposed …

This paper presents CORE, a conceptually simple, effective and communication-efficient
model for multi-agent cooperative perception. It addresses the task from a novel perspective …

In this paper, we introduce a new self-supervised rationalization method, called KGRec, for
knowledge-aware recommender systems. To effectively identify informative knowledge …