Action Detection via an Image Diffusion Process

Foo, Lin Geng and Li, Tianjiao and Rahmani, Hossein and Liu, Jun (2024) Action Detection via an Image Diffusion Process. Other. UNSPECIFIED.

[thumbnail of 2404.01051v1]
Text (2404.01051v1)
Download (0B)
[thumbnail of 2404.01051v1]
Text (2404.01051v1)
Download (0B)
[thumbnail of 2404.01051v1]
Text (2404.01051v1)
Download (0B)
[thumbnail of 2404.01051v1]
Text (2404.01051v1)
Download (0B)
[thumbnail of 2404.01051v1]
Text (2404.01051v1)
Download (0B)
[thumbnail of 2404.01051v1]
Text (2404.01051v1)
Download (0B)
[thumbnail of 2404.01051v1]
Text (2404.01051v1)
Download (0B)
[thumbnail of 2404.01051v1]
Text (2404.01051v1)
Download (0B)
[thumbnail of 2404.01051v1]
Text (2404.01051v1)
Download (0B)
[thumbnail of 2404.01051v1]
Text (2404.01051v1)
Download (0B)
[thumbnail of 2404.01051v1]
Text (2404.01051v1)
Download (0B)
[thumbnail of 2404.01051v1]
Text (2404.01051v1)
2404.01051v1.pdf

Download (803kB)

Abstract

Action detection aims to localize the starting and ending points of action instances in untrimmed videos, and predict the classes of those instances. In this paper, we make the observation that the outputs of the action detection task can be formulated as images. Thus, from a novel perspective, we tackle action detection via a three-image generation process to generate starting point, ending point and action-class predictions as images via our proposed Action Detection Image Diffusion (ADI-Diff) framework. Furthermore, since our images differ from natural images and exhibit special properties, we further explore a Discrete Action-Detection Diffusion Process and a Row-Column Transformer design to better handle their processing. Our ADI-Diff framework achieves state-of-the-art results on two widely-used datasets.

Item Type:
Monograph (Other)
Additional Information:
Accepted to CVPR 2024
Subjects:
?? cs.cv ??
ID Code:
221075
Deposited By:
Deposited On:
11 Jun 2024 12:35
Refereed?:
No
Published?:
Published
Last Modified:
15 Jul 2024 08:02