SOVC: Subject-Oriented Video Captioning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Teng, Chang, Ma, Yunchuan, Li, Guorong, Qi, Yuankai, Qing, Laiyu, Huang, Qingming
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909308191506432
author Teng, Chang
Ma, Yunchuan
Li, Guorong
Qi, Yuankai
Qing, Laiyu
Huang, Qingming
author_facet Teng, Chang
Ma, Yunchuan
Li, Guorong
Qi, Yuankai
Qing, Laiyu
Huang, Qingming
contents Describing video content according to users' needs is a long-held goal. Although existing video captioning methods have made significant progress, the generated captions may not focus on the entity that users are particularly interested in. To address this problem, we propose a new video captioning task, Subject-Oriented Video Captioning (SOVC), which aims to allow users to specify the describing target via a bounding box. To support this task, we construct two subject-oriented video captioning datasets based on two widely used video captioning datasets: MSVD and MSRVTT, by annotating subjects in each video for each caption. These datasets pave the way for describing users' interested targets. To tackle this task, we introduce a method tailored to this task, named SOVCNet. It consists of two key components: a subject-oriented sampling module that samples frames related to the subject to minimize irrelevant information; and a subject-oriented encoding module that utilizes the subject areas as hard prompts and integrates learnable soft prompts, enhancing the model's focus on the subject's activities and facilitating adaptation to the downstream generation task. Extensive experimental results demonstrate the effectiveness of our method on this new task.
format Preprint
id arxiv_https___arxiv_org_abs_2312_13330
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle SOVC: Subject-Oriented Video Captioning
Teng, Chang
Ma, Yunchuan
Li, Guorong
Qi, Yuankai
Qing, Laiyu
Huang, Qingming
Computer Vision and Pattern Recognition
Describing video content according to users' needs is a long-held goal. Although existing video captioning methods have made significant progress, the generated captions may not focus on the entity that users are particularly interested in. To address this problem, we propose a new video captioning task, Subject-Oriented Video Captioning (SOVC), which aims to allow users to specify the describing target via a bounding box. To support this task, we construct two subject-oriented video captioning datasets based on two widely used video captioning datasets: MSVD and MSRVTT, by annotating subjects in each video for each caption. These datasets pave the way for describing users' interested targets. To tackle this task, we introduce a method tailored to this task, named SOVCNet. It consists of two key components: a subject-oriented sampling module that samples frames related to the subject to minimize irrelevant information; and a subject-oriented encoding module that utilizes the subject areas as hard prompts and integrates learnable soft prompts, enhancing the model's focus on the subject's activities and facilitating adaptation to the downstream generation task. Extensive experimental results demonstrate the effectiveness of our method on this new task.
title SOVC: Subject-Oriented Video Captioning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2312.13330