Tu-Khiem Le
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Tu-Khiem Le has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Minh-Triet Tran
University of Science
12 shared publications - Cathal Gurrin
Dublin City University, Dublin, Ireland
10 shared publications - Van-Tu Ninh
Dublin City University
10 shared publications - Minh-Khoi Pham
Dublin City University
6 shared publications - Duc-Tien Dang-Nguyen
University of Bergen, Bergen, Norway
5 shared publications - Gia-Huy Vuong
University of Science, VNU-HCM
5 shared publications
Latest Publications
- Top Cited Papers
- 2 September 2026
- Published by Springer Nature
- p. 124-138
Toward Abstraction-Level Event Retrieval in Large Video Collections: Leveraging Human Knowledge and LLM-Based Reasoning in the Ho Chi Minh City AI Challenge 2025
- 2 September 2026
- Published by Springer Nature
- p. 124-138
Top Cited PapersToward Abstraction-Level Event Retrieval in Large Video Collections: Leveraging Human Knowledge and LLM-Based Reasoning in the Ho Chi Minh City AI Challenge 2025
- 26 April 2025
- Published by Springer Nature
- p. 3-17
Event Retrieval from Large Video Collection in Ho Chi Minh City AI Challenge 2024
- 26 April 2025
- Published by Springer Nature
- p. 3-17
Event Retrieval from Large Video Collection in Ho Chi Minh City AI Challenge 2024
- 1 January 2025
- Published by Springer Nature
- p. 371-377
ViewsInsight2.0: Enhancing Video Retrieval for VBS 2025 with an Automatic Query Generator Powered by Large Language Models
- 1 January 2025
- Published by Springer Nature
- p. 371-377
ViewsInsight2.0: Enhancing Video Retrieval for VBS 2025 with an Automatic Query Generator Powered by Large Language Models
- Open Access
- 10 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
LifeInsight2.0: An Enhanced Approach for Automated Lifelog Retrieval in LSC'24
We introduce the LifeInsight 2.0 system - an enhanced version of LifeInsight, built specifically for the sixth annual Lifelog Search Challenge (LSC'23). LifeInsight 2.0 leverages the core functionalities of LifeInsight while incorporating significant improvements to address performance bottlenecks. This refined architecture aims to deliver superior search capabilities within the LSC'24. LifeInsight 2.0 employs an ensemble approach combining two powerful foundation models: CLIP (Contrastive Language-Image Pretraining) and BLIP2 (Bootstrapping Language-Image Pretraining) model. In addition, the system incorporates a temporal query mechanism and an automatic query parser. The former enables LifeInsight 2.0 to interpret queries that include time-based information, while the latter specifically handles tasks involving question answering.
- 10 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessLifeInsight2.0: An Enhanced Approach for Automated Lifelog Retrieval in LSC'24
We introduce the LifeInsight 2.0 system - an enhanced version of LifeInsight, built specifically for the sixth annual Lifelog Search Challenge (LSC'23). LifeInsight 2.0 leverages the core functionalities of LifeInsight while incorporating significant improvements to address performance bottlenecks. This refined architecture aims to deliver superior search capabilities within the LSC'24. LifeInsight 2.0 employs an ensemble approach combining two powerful foundation models: CLIP (Contrastive Language-Image Pretraining) and BLIP2 (Bootstrapping Language-Image Pretraining) model. In addition, the system incorporates a temporal query mechanism and an automatic query parser. The former enables LifeInsight 2.0 to interpret queries that include time-based information, while the latter specifically handles tasks involving question answering.
- Open Access
- 10 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 53-57
LifeSeeker 6.0: Leveraging the linguistic aspect of the lifelog system in LSC'24
Supporting effective access to digital lifelogs is a challenging research task because of both the volume and variety of multimodal lifelog data, as well as the many and diverse types of information need that should be supported. In this paper, we introduce a new version of LifeSeeker called LifeSeeker 6.0, for the 2024 edition of the ACM Lifelog Search Challenge. Our enhancements include the improvements to the user interface and the backend reconstruction by combining the E-LifeSeeker structure with using contrastive learning between texts. These adjustments are aimed at accelerating the correlation between the huge image collection and the text input, thereby enhancing the retrieval accuracy and efficiency.
- 10 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 53-57
Open AccessLifeSeeker 6.0: Leveraging the linguistic aspect of the lifelog system in LSC'24
Supporting effective access to digital lifelogs is a challenging research task because of both the volume and variety of multimodal lifelog data, as well as the many and diverse types of information need that should be supported. In this paper, we introduce a new version of LifeSeeker called LifeSeeker 6.0, for the 2024 edition of the ACM Lifelog Search Challenge. Our enhancements include the improvements to the user interface and the backend reconstruction by combining the E-LifeSeeker structure with using contrastive learning between texts. These adjustments are aimed at accelerating the correlation between the huge image collection and the text input, thereby enhancing the retrieval accuracy and efficiency.
- Open Access
- 10 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 41-46
LifeLens 2.0: Improving Efficiency and Usability in Lifelog Retrieval Systems through UX/UI Design
In lifelogging, user experience is one of the most important features as it provides accessibility and efficiency. A well-designed and intuitive interface minimizes errors and frustration, which provides the ability to find a specific lifelog image quickly and easily. Although lifelogging is well-established for retrieving information, there's still much to explore in improving user experience. This paper presents the development of LifeLens 2.0, a lifelogging system designed to capture, organize, and retrieve personal life events through advanced wearable technology. LifeLens introduces significant enhancements over LSC'24. By adopting a minimalist yet vibrant design philosophy, we have also redefined the user interface to be more appealing and intuitive. These enhancements demonstrate our dedication to merging user-friendly design with efficiency, catering to both novice and expert users in the lifelogging field.
- 10 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 41-46
Open AccessLifeLens 2.0: Improving Efficiency and Usability in Lifelog Retrieval Systems through UX/UI Design
In lifelogging, user experience is one of the most important features as it provides accessibility and efficiency. A well-designed and intuitive interface minimizes errors and frustration, which provides the ability to find a specific lifelog image quickly and easily. Although lifelogging is well-established for retrieving information, there's still much to explore in improving user experience. This paper presents the development of LifeLens 2.0, a lifelogging system designed to capture, organize, and retrieve personal life events through advanced wearable technology. LifeLens introduces significant enhancements over LSC'24. By adopting a minimalist yet vibrant design philosophy, we have also redefined the user interface to be more appealing and intuitive. These enhancements demonstrate our dedication to merging user-friendly design with efficiency, catering to both novice and expert users in the lifelogging field.
- 29 January 2024
- Published by Springer Nature
- p. 400-406
ViewsInsight: Enhancing Video Retrieval for VBS 2024 with a User-Friendly Interaction Mechanism
- 29 January 2024
- Published by Springer Nature
- p. 400-406
ViewsInsight: Enhancing Video Retrieval for VBS 2024 with a User-Friendly Interaction Mechanism
- Open Access
- 7 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 893-900
NewsInsight: A Comprehensive Video Event Retrieval System with Spatial Insights and Query Assistance
Video event retrieval is the task of finding videos that are relevant to a given query. It is a challenging problem because videos are typically much larger than images, and they can contain a variety of different objects and scenes. However, there are a number of different approaches to video retrieval, and the field is rapidly evolving. Some of the most promising research directions include the use of deep learning and multimodal features. In this paper, we introduce NewsInsight – a comprehensive video event retrieval system developed for participating AI Challenge 2023. The system under investigation leverages the Bootstrapping Language-Image Pre-training (BLIP) model for zero-shot image-text retrieval, demonstrating superior recall scores on the Flickr30K dataset compared to the Contrastive Language–Image Pretraining (CLIP) model. In addition, it employs an Elastic Search filtering mechanism to discard irrelevant images. Beyond semantic search mechanisms, the system supports visual similarity search by calculating the inner product distance between vectors in the video frames corpus and the query image. The system also incorporates an explicit relevance feedback function, AI-based query description rewriting, and visual-example-generating features, enhancing the precision of the query description and aiding end-users in formulating a more accurate depiction of the targeted image for retrieval.
- 7 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 893-900
Open AccessNewsInsight: A Comprehensive Video Event Retrieval System with Spatial Insights and Query Assistance
Video event retrieval is the task of finding videos that are relevant to a given query. It is a challenging problem because videos are typically much larger than images, and they can contain a variety of different objects and scenes. However, there are a number of different approaches to video retrieval, and the field is rapidly evolving. Some of the most promising research directions include the use of deep learning and multimodal features. In this paper, we introduce NewsInsight – a comprehensive video event retrieval system developed for participating AI Challenge 2023. The system under investigation leverages the Bootstrapping Language-Image Pre-training (BLIP) model for zero-shot image-text retrieval, demonstrating superior recall scores on the Flickr30K dataset compared to the Contrastive Language–Image Pretraining (CLIP) model. In addition, it employs an Elastic Search filtering mechanism to discard irrelevant images. Beyond semantic search mechanisms, the system supports visual similarity search by calculating the inner product distance between vectors in the video frames corpus and the query image. The system also incorporates an explicit relevance feedback function, AI-based query description rewriting, and visual-example-generating features, enhancing the precision of the query description and aiding end-users in formulating a more accurate depiction of the targeted image for retrieval.
- Open Access
- 7 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 1011-1017
News Event Retrieval from Large Video Collection in Ho Chi Minh City AI Challenge 2023
Event retrieval from large collections of TV news videos is crucial for efficient information access, enabling researchers, journalists, and the general public to quickly locate and analyze relevant content amidst the vast sea of news coverage, facilitating informed decision-making and a comprehensive understanding of significant events. This paper presents an overview of the AI-driven video retrieval task in Ho Chi Minh City AI Challenge 2023. The competition draws inspiration from internationally recognized competitions, namely the Video Browser Showdown (VBS) and the Lifelog Search Challenge (LSC). Participants are tasked with developing AI models to retrieve specific video segments from a diverse dataset from reputable news channels. The dataset comprises a vast collection of videos, keyframes, object detections, CLIP features, and metadata. It is divided into three packs with a total of 1,270 videos, spanning approximately 360 hours of content. The challenge comprises two groups. Group A is open to students, researchers, and practitioners in artificial intelligence and information retrieval, emphasizing substantial knowledge and experience. Group B is tailored for high school students, focusing on nurturing interest, learning, and engagement among the next generation of AI enthusiasts. The wide variation in the content of queries challenged participants to demonstrate their adaptability and creativity in effectively retrieving diverse events from the extensive TV news video dataset. The winning teams showcased promising solutions by effectively harnessing artificial intelligence and information retrieval techniques to excel in event retrieval from a vast collection of TV news videos.
- 7 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 1011-1017
Open AccessNews Event Retrieval from Large Video Collection in Ho Chi Minh City AI Challenge 2023
Event retrieval from large collections of TV news videos is crucial for efficient information access, enabling researchers, journalists, and the general public to quickly locate and analyze relevant content amidst the vast sea of news coverage, facilitating informed decision-making and a comprehensive understanding of significant events. This paper presents an overview of the AI-driven video retrieval task in Ho Chi Minh City AI Challenge 2023. The competition draws inspiration from internationally recognized competitions, namely the Video Browser Showdown (VBS) and the Lifelog Search Challenge (LSC). Participants are tasked with developing AI models to retrieve specific video segments from a diverse dataset from reputable news channels. The dataset comprises a vast collection of videos, keyframes, object detections, CLIP features, and metadata. It is divided into three packs with a total of 1,270 videos, spanning approximately 360 hours of content. The challenge comprises two groups. Group A is open to students, researchers, and practitioners in artificial intelligence and information retrieval, emphasizing substantial knowledge and experience. Group B is tailored for high school students, focusing on nurturing interest, learning, and engagement among the next generation of AI enthusiasts. The wide variation in the content of queries challenged participants to demonstrate their adaptability and creativity in effectively retrieving diverse events from the extensive TV news video dataset. The winning teams showcased promising solutions by effectively harnessing artificial intelligence and information retrieval techniques to excel in event retrieval from a vast collection of TV news videos.
- 21 September 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 82 (24) , 37877-37902
First-flexible interactive retrieval system for visual lifelog exploration
- 21 September 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 82 (24) , 37877-37902
First-flexible interactive retrieval system for visual lifelog exploration