Zobrazeno 1 - 10
of 191
pro vyhledávání: '"Jiang Zhongyu"'
The Segment Anything Model 2 (SAM 2) has demonstrated strong performance in object segmentation tasks but faces challenges in visual object tracking, particularly when managing crowded scenes with fast-moving or self-occluding objects. Furthermore, t
Externí odkaz:
http://arxiv.org/abs/2411.11922
Multi-object tracking in sports scenarios has become one of the focal points in computer vision, experiencing significant advancements through the integration of deep learning techniques. Despite these breakthroughs, challenges remain, such as accura
Externí odkaz:
http://arxiv.org/abs/2411.08216
Autor:
Cai, Chengkun, Zhao, Xu, Liu, Haoliang, Jiang, Zhongyu, Zhang, Tianfang, Wu, Zongkai, Hwang, Jenq-Neng, Li, Lei
Large Language Models (LLMs) have achieved substantial progress in artificial intelligence, particularly in reasoning tasks. However, their reliance on static prompt structures, coupled with limited dynamic reasoning capabilities, often constrains th
Externí odkaz:
http://arxiv.org/abs/2410.02892
Autor:
Huang, Hsiang-Wei, Sun, Jiacheng, Yang, Cheng-Yen, Jiang, Zhongyu, Huang, Li-Yu, Hwang, Jenq-Neng, Yeh, Yu-Ching
Assessing gross motor development in toddlers is crucial for understanding their physical development and identifying potential developmental delays or disorders. However, existing datasets for action recognition primarily focus on adults, lacking th
Externí odkaz:
http://arxiv.org/abs/2409.00349
Autor:
Chen, Shen, Zhou, Jiale, Jiang, Zhongyu, Zhang, Tianfang, Wu, Zongkai, Hwang, Jenq-Neng, Li, Lei
The creation of high-quality 3D assets is paramount for applications in digital heritage preservation, entertainment, and robotics. Traditionally, this process necessitates skilled professionals and specialized software for the modeling, texturing, a
Externí odkaz:
http://arxiv.org/abs/2407.19035
Autor:
Ho, Yuan-Hao, Cheng, Jen-Hao, Kuan, Sheng Yao, Jiang, Zhongyu, Chai, Wenhao, Huang, Hsiang-Wei, Lin, Chih-Lung, Hwang, Jenq-Neng
Traditional methods for human localization and pose estimation (HPE), which mainly rely on RGB images as an input modality, confront substantial limitations in real-world applications due to privacy concerns. In contrast, radar-based HPE methods emer
Externí odkaz:
http://arxiv.org/abs/2407.13930
In the field of multi-object tracking (MOT), traditional methods often rely on the Kalman Filter for motion prediction, leveraging its strengths in linear motion scenarios. However, the inherent limitations of these methods become evident when confro
Externí odkaz:
http://arxiv.org/abs/2403.10826
Autor:
Yang, Cheng-Yen, Huang, Hsiang-Wei, Jiang, Zhongyu, Wang, Hao, Wallace, Farron, Hwang, Jenq-Neng
Dense object counting or crowd counting has come a long way thanks to the recent development in the vision community. However, indiscernible object counting, which aims to count the number of targets that are blended with respect to their surrounding
Externí odkaz:
http://arxiv.org/abs/2403.03461
Autor:
Li, Lei, Zhang, Tianfang, Jiang, Zhongyu, Yang, Cheng-Yen, Hwang, Jenq-Neng, Oehmcke, Stefan, Gominski, Dimitri Pierre Johannes, Gieseke, Fabian, Igel, Christian
Accurate and consistent methods for counting trees based on remote sensing data are needed to support sustainable forest management, assess climate change mitigation strategies, and build trust in tree carbon credits. Two-dimensional remote sensing i
Externí odkaz:
http://arxiv.org/abs/2403.01932
Although 3D human pose estimation has gained impressive development in recent years, only a few works focus on infants, that have different bone lengths and also have limited data. Directly applying adult pose estimation models typically achieves low
Externí odkaz:
http://arxiv.org/abs/2311.12043