Duy-Khang Ho
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Duy-Khang Ho has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Minh-Quan Ho-Le
University of Science, VNU-HCM, Ho Chi Minh City, Vietnam and Viet Nam National University, Ho Chi Minh City, Vietnam
9 shared publications - Minh-Triet Tran
University of Science
9 shared publications - Van-Tu Ninh
Dublin City University
6 shared publications - Huy-Hoang Do-Huu
VNU-HCM
5 shared publications - Nhut-Thanh Le-Hinh
University of Science, VNU-HCM
5 shared publications - Cathal Gurrin
Dublin City University, Dublin, Ireland
4 shared publications
Latest Publications
- 12 August 2026
- Published by Springer Nature
- p. 510-524
Applying Large Language Model (LLM) Agents for Automated Lifelog Retrieval
- 12 August 2026
- Published by Springer Nature
- p. 510-524
Applying Large Language Model (LLM) Agents for Automated Lifelog Retrieval
- 7 February 2026
- Published by Springer Nature
- p. 191-198
From Expert Practices to Intelligent Agents: Autonomy in Interactive Video Retrieval
- 7 February 2026
- Published by Springer Nature
- p. 191-198
From Expert Practices to Intelligent Agents: Autonomy in Interactive Video Retrieval
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12600-12606
LSC-ADL: An Activity of Daily Living (ADL)-Annotated Lifelog Dataset Generated via Semi-Automatic Clustering
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12600-12606
Open AccessLSC-ADL: An Activity of Daily Living (ADL)-Annotated Lifelog Dataset Generated via Semi-Automatic Clustering
- Open Access
- 30 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 7-14
SnapSeek 3.0: Interactive Lifelog Search System with Enhanced Scene Understanding
- 30 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 7-14
Open AccessSnapSeek 3.0: Interactive Lifelog Search System with Enhanced Scene Understanding
- Open Access
- 30 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 15-22
LUMINA-1: Learning and Understanding Multimedia in Immersive Navigable Archives for Lifelog Retrieval
- 30 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 15-22
Open AccessLUMINA-1: Learning and Understanding Multimedia in Immersive Navigable Archives for Lifelog Retrieval
- 26 April 2025
- Published by Springer Nature
- p. 280-293
SnapSeek: A Multimodal Video Retrieval System with Context Awareness for AI Challenge 2024
- 26 April 2025
- Published by Springer Nature
- p. 280-293
SnapSeek: A Multimodal Video Retrieval System with Context Awareness for AI Challenge 2024
- 26 April 2025
- Published by Springer Nature
- p. 326-336
Media Certificate Authority: A System to Ensure Media Content Originality for Daily Lifelog Media Collection
- 26 April 2025
- Published by Springer Nature
- p. 326-336
Media Certificate Authority: A System to Ensure Media Content Originality for Daily Lifelog Media Collection
- Open Access
- 10 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 24-29
SnapSeek: An Interactive Lifelog Acquisition System for LSC'24
In the digital age, the surge in lifelog data presents challenges for traditional search methods due to their reliance on natural language and lack of context. Our research introduces an advanced semantic search application tailored to Lifelog Search Challenge 2024. Leveraging CLIP, BLIP-2, BEIT-3, and all-mpnet-base-v2 models, our system creates precise embeddings from diverse data formats. With Milvus, swift and accurate data retrieval is ensured for efficient indexing. Unique features include metadata enhancement, query standardization, query auto-parsing, effective ranking system, and innovative UI for further lifelog exploration. These advancements redefine lifelog search, offering enhanced access and navigation within archives, and setting a new standard for intelligent, context-aware solutions.
- 10 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 24-29
Open AccessSnapSeek: An Interactive Lifelog Acquisition System for LSC'24
In the digital age, the surge in lifelog data presents challenges for traditional search methods due to their reliance on natural language and lack of context. Our research introduces an advanced semantic search application tailored to Lifelog Search Challenge 2024. Leveraging CLIP, BLIP-2, BEIT-3, and all-mpnet-base-v2 models, our system creates precise embeddings from diverse data formats. With Milvus, swift and accurate data retrieval is ensured for efficient indexing. Unique features include metadata enhancement, query standardization, query auto-parsing, effective ranking system, and innovative UI for further lifelog exploration. These advancements redefine lifelog search, offering enhanced access and navigation within archives, and setting a new standard for intelligent, context-aware solutions.
- Open Access
- 7 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 435-442
Improving Bird Sounds Classification with Additional Fine-tuning on Augmented Data
Many bird species are currently on the verge of extinction and functionally extinct. Therefore, measures to note down the quantity and distribution of bird species using AI models have always been in consideration to support the biodiversity monitoring process. This study aims to enhance the accuracy of bird sound identification by applying fine-tuning with data augmentation to the existing CNN-based model. On the dataset from the BirdCLEF 2021 competition, the model subsequently achieves the F1-score of 0.6928 in the private set and 0.7566 in the public set, increasing noticeably by 0.0465 and 0.0991 respectively compared to the rebuilt bird classification solution proposed by Naoki Murakami, Hajime Tanaka and Masataka Nishimori, and has the ability to predict nearly 400 species of birds more accurately. The improved performance of the model is expected to support the efforts of conservationists in monitoring bird distribution and promoting species preservation.
- 7 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 435-442
Open AccessImproving Bird Sounds Classification with Additional Fine-tuning on Augmented Data
Many bird species are currently on the verge of extinction and functionally extinct. Therefore, measures to note down the quantity and distribution of bird species using AI models have always been in consideration to support the biodiversity monitoring process. This study aims to enhance the accuracy of bird sound identification by applying fine-tuning with data augmentation to the existing CNN-based model. On the dataset from the BirdCLEF 2021 competition, the model subsequently achieves the F1-score of 0.6928 in the private set and 0.7566 in the public set, increasing noticeably by 0.0465 and 0.0991 respectively compared to the rebuilt bird classification solution proposed by Naoki Murakami, Hajime Tanaka and Masataka Nishimori, and has the ability to predict nearly 400 species of birds more accurately. The improved performance of the model is expected to support the efforts of conservationists in monitoring bird distribution and promoting species preservation.