Evaluation of Winning Solutions of 2025 Low Power Computer Vision Challenge
Zihao Ye, Yung-Hsiang Lu, Xiao Hu, Shuai Zhang, Taotao Jing, Xin Li, Zhen Yao, Bo Lang, Zhihao Zheng, Seungmin Oh, Hankyul Kang, Seunghun Kang, Jongbin Ryu, Kexin Chen, Yuan Qi, George K Thiruvathukal, and Mooi Choo Chuah
TL;DR
This paper reviews the design, evaluation framework, and top solutions of the 2025 Low Power Computer Vision Challenge, emphasizing efficient models for edge devices across three vision tasks.
Contribution
It introduces the LPCVC 2025 competition structure, evaluation methods, and highlights innovative solutions and trends in low-power computer vision.
Findings
Top solutions demonstrate improved efficiency and accuracy.
Evaluation framework ensures reproducibility across devices.
Key trends include model compression and multi-task learning.
Abstract
The IEEE Low-Power Computer Vision Challenge (LPCVC) aims to promote the development of efficient vision models for edge devices, balancing accuracy with constraints such as latency, memory capacity, and energy use. The 2025 challenge featured three tracks: (1) Image classification under various lighting conditions and styles, (2) Open-Vocabulary Segmentation with Text Prompt, and (3) Monocular Depth Estimation. This paper presents the design of LPCVC 2025, including its competition structure and evaluation framework, which integrates the Qualcomm AI Hub for consistent and reproducible benchmarking. The paper also introduces the top-performing solutions from each track and outlines key trends and observations. The paper concludes with suggestions for future computer vision competitions.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.