1 code implementation • 18 Oct 2023 • Shaoxiong Duan, Yining Shi, Wei Xu
We then introduce Attention Bias Calibration (ABC), a calibration stage that enables the model to automatically learn the proper attention biases, which we show to be connected to mechanisms in relative position encoding.
no code implementations • 2 Mar 2023 • Yining Shi, Kun Jiang, Jiusi Li, Junze Wen, Zelin Qian, Mengmeng Yang, Ke Wang, Diange Yang
Grid-centric perception is a crucial field for mobile robot perception and navigation.
1 code implementation • 19 Feb 2023 • Yining Shi, Kun Jiang, Ke Wang, Jiusi Li, Yunlong Wang, Diange Yang
This paper investigates multi-sensor spatio-temporal fusion strategies for continuous occupancy prediction in a systematic manner.
no code implementations • 25 Aug 2022 • Taohua Zhou, Yining Shi, Junjie Chen, Kun Jiang, Mengmeng Yang, Diange Yang
A novel method that realizes the feature-level fusion under the bird's-eye view (BEV) for a better feature representation is proposed.
2 code implementations • 29 Jun 2022 • Yining Shi, Jingyan Shen, Yifan Sun, Yunlong Wang, Jiaxin Li, Shiqi Sun, Kun Jiang, Diange Yang
Our novel sparse feature sampling module only utilizes local 2D region of interest (RoI) features calculated by the projection of 3D query boxes for further box refinement, leading to a fully-convolutional and deployment-friendly pipeline.
Ranked #8 on Robust Camera Only 3D Object Detection on nuScenes-C
no code implementations • 17 Dec 2021 • David Chuan-En Lin, Anastasis Germanidis, Cristóbal Valenzuela, Yining Shi, Nikolas Martelaro
In the art of video editing, sound helps add character to an object and immerse the viewer within a space.
3 code implementations • 14 Dec 2021 • Xupeng Miao, Hailin Zhang, Yining Shi, Xiaonan Nie, Zhi Yang, Yangyu Tao, Bin Cui
Embedding models have been an effective learning paradigm for high-dimensional data.