no code implementations • 23 Sep 2023 • Guotian Zeng, Bi Zeng, Hong Zhang, Jianqi Liu, Qingmao Wei
Single object tracking (SOT) heavily relies on the representation of the target object as a bounding box.
1 code implementation • 17 Sep 2023 • Qingmao Wei, Bi Zeng, Jianqi Liu, Li He, Guotian Zeng
As an example, our fastest variant, LiteTrack-B4, achieves 65. 2% AO on the GOT-10k benchmark, surpassing all preceding efficient trackers, while running over 100 fps with ONNX on the Jetson Orin NX edge device.
no code implementations • 6 Sep 2023 • Qingmao Wei, Bi Zeng, Guotian Zeng
This study introduces a more efficient training strategy to mitigate overfitting and reduce computational requirements.
no code implementations • 6 Sep 2023 • Qingmao Wei, Guotian Zeng, Bi Zeng
Recent Transformer-based visual tracking models have showcased superior performance.
no code implementations • 9 Feb 2021 • Pengfei Wei, Bi Zeng, Wenxiong Liao
In this paper, we propose a new joint model with a wheel-graph attention network (Wheel-GAT) which is able to model interrelated connections directly for intent detection and slot filling.