IEEE

Light-SLAM: A Robust Deep-Learning Visual SLAM System Based on LightGlue Under Challenging Lighting Conditions

Abstract: Simultaneous Localization and Mapping (SLAM) has become a critical technology for intelligent transportation systems and autonomous robots and is widely used in autonomous driving. However, ...
Princeton University

In the hallways of Princeton, a fascination with the human mind unlocked the power of deep learning

Serendipitous meetings, scholarly collaborations, and an ethos of "encouraging junior faculty to think big" laid the ...
IEEE

Deep Learning-Based Approaches for Enhanced Image Compression in Computer Vision Applications

Abstract: Image compression is required for many computer vision applications, but traditional methods like JPEG and PNG have trouble preserving image quality at high compression levels. The study ...
The Lancet

Causal deep learning for real-time detection of cardiac surgery-associated acute kidney injury: derivation and validation in seven time-series cohorts

aMedical Big Data Research Center, Chinese The People’s Liberation Army General Hospital, Beijing, China bNational Engineering Research Center of Medical Big Data Application Technology, The People’s ...
GitHub

[Bug] Use Qwen2.5-VL-3B-Insturct for Vision Reinforcement Learning

[rank0]: Traceback (most recent call last): [rank0]: File "/home/wangkeyi/.conda/envs/rl/lib/python3.12/site-packages/unsloth_zoo/vllm_utils.py", line 1665, in load ...