Enhancing performance in video grounding tasks through the use of attention module

This report investigates improving video grounding tasks through the use of attention mechanisms, tackling the issue of sparse annotations in video datasets. Drawing inspiration from the MMN model \cite{wang2021_negative_2dmap}, we developed a modified model based on the open-source MMN codebase and...

Full description

Saved in:
Bibliographic Details
Main Author: Do Duc Anh
Other Authors: Sun Aixin
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2024
Subjects:
Online Access:https://hdl.handle.net/10356/181703
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-181703
record_format dspace
spelling sg-ntu-dr.10356-1817032024-12-16T01:26:13Z Enhancing performance in video grounding tasks through the use of attention module Do Duc Anh Sun Aixin College of Computing and Data Science AXSun@ntu.edu.sg Computer and Information Science This report investigates improving video grounding tasks through the use of attention mechanisms, tackling the issue of sparse annotations in video datasets. Drawing inspiration from the MMN model \cite{wang2021_negative_2dmap}, we developed a modified model based on the open-source MMN codebase and evaluated it on several widely-used datasets, including Charades-STA and ActivityNet Captions. Our approach shows improvements over certain benchmarks. Additionally, we conducted an in-depth analysis to assess the role of attention in enhancing the multimodal framework's ability to comprehend the complex structure of videos. Bachelor's degree 2024-12-16T01:26:13Z 2024-12-16T01:26:13Z 2024 Final Year Project (FYP) Do Duc Anh (2024). Enhancing performance in video grounding tasks through the use of attention module. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/181703 https://hdl.handle.net/10356/181703 en application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Computer and Information Science
spellingShingle Computer and Information Science
Do Duc Anh
Enhancing performance in video grounding tasks through the use of attention module
description This report investigates improving video grounding tasks through the use of attention mechanisms, tackling the issue of sparse annotations in video datasets. Drawing inspiration from the MMN model \cite{wang2021_negative_2dmap}, we developed a modified model based on the open-source MMN codebase and evaluated it on several widely-used datasets, including Charades-STA and ActivityNet Captions. Our approach shows improvements over certain benchmarks. Additionally, we conducted an in-depth analysis to assess the role of attention in enhancing the multimodal framework's ability to comprehend the complex structure of videos.
author2 Sun Aixin
author_facet Sun Aixin
Do Duc Anh
format Final Year Project
author Do Duc Anh
author_sort Do Duc Anh
title Enhancing performance in video grounding tasks through the use of attention module
title_short Enhancing performance in video grounding tasks through the use of attention module
title_full Enhancing performance in video grounding tasks through the use of attention module
title_fullStr Enhancing performance in video grounding tasks through the use of attention module
title_full_unstemmed Enhancing performance in video grounding tasks through the use of attention module
title_sort enhancing performance in video grounding tasks through the use of attention module
publisher Nanyang Technological University
publishDate 2024
url https://hdl.handle.net/10356/181703
_version_ 1819113015852662784