Contrastive Learning for Automotive mmWave Radar Detection Points Based Instance Segmentation

被引:11
|
作者
Xiong, Weiyi [1 ]
Liu, Jianan [2 ,3 ]
Xia, Yuxuan [4 ]
Huang, Tao [5 ]
Zhu, Bing [1 ]
Xiang, Wei [6 ]
机构
[1] Beihang Univ, Sch Automat Sci & Elect Engn, Beijing 100191, Peoples R China
[2] Vitalent Consulting, S-41761 Gothenburg, Sweden
[3] Silo AI, Stockholm, Sweden
[4] Chalmers Univ Technol, Dept Elect Engn, S-41296 Gothenburg, Sweden
[5] James Cook Univ, Coll Sci & Engn, Cairns, Australia
[6] La Trobe Univ, Sch Engn & Math Sci, Melbourne, Vic, Australia
关键词
D O I
10.1109/ITSC55140.2022.9922540
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The automotive mmWave radar plays a key role in advanced driver assistance systems (ADAS) and autonomous driving. Deep learning-based instance segmentation enables real-time object identification from the radar detection points. In the conventional training process, accurate annotation is the key. However, high-quality annotations of radar detection points are challenging to achieve due to their ambiguity and sparsity. To address this issue, we propose a contrastive learning approach for implementing radar detection points-based instance segmentation. We define the positive and negative samples according to the ground-truth label, apply the contrastive loss to train the model first, and then perform fine-tuning for the following downstream task. In addition, these two steps can be merged into one, and pseudo labels can be generated for the unlabeled data to improve the performance further. Thus, there are four different training settings for our method. Experiments show that when the ground-truth information is only available for a small proportion of the training data, our method still achieves a comparable performance to the approach trained in a supervised manner with 100% ground-truth information.
引用
收藏
页码:1255 / 1261
页数:7
相关论文
共 50 条
  • [21] Deep Learning Based Road Boundary Detection Using Camera and Automotive Radar
    Patel, Dipkumar
    Elgazzar, Khalid
    2024 35TH IEEE INTELLIGENT VEHICLES SYMPOSIUM, IEEE IV 2024, 2024, : 3235 - 3240
  • [22] Semantic Segmentation-Based Occupancy Grid Map Learning With Automotive Radar Raw Data
    Jin, Yi
    Hoffmann, Marcel
    Deligiannis, Anastasios
    Fuentes-Michel, Juan-Carlos
    Vossiek, Martin
    IEEE TRANSACTIONS ON INTELLIGENT VEHICLES, 2024, 9 (01): : 216 - 230
  • [23] Detection and Tracking on Automotive Radar Data with Deep Learning
    Tilly, Julius F.
    Haag, Stefan
    Schumann, Ole
    Weishaupt, Fabio
    Duraisamy, Bharanidhar
    Dickmann, Jurgen
    Fritzsche, Martin
    PROCEEDINGS OF 2020 23RD INTERNATIONAL CONFERENCE ON INFORMATION FUSION (FUSION 2020), 2020, : 1028 - 1034
  • [24] A deep learning approach for insulator instance segmentation and defect detection
    Antwi-Bekoe, Eldad
    Liu, Guisong
    Ainam, Jean-Paul
    Sun, Guolin
    Xie, Xiurui
    NEURAL COMPUTING & APPLICATIONS, 2022, 34 (09): : 7253 - 7269
  • [25] A deep learning approach for insulator instance segmentation and defect detection
    Eldad Antwi-Bekoe
    Guisong Liu
    Jean-Paul Ainam
    Guolin Sun
    Xiurui Xie
    Neural Computing and Applications, 2022, 34 : 7253 - 7269
  • [26] Hierarchical-Instance Contrastive Learning for Minority Detection on Imbalanced Medical Datasets
    Li, Yiyue
    Qian, Guangwu
    Jiang, Xiaoshuang
    Jiang, Zekun
    Wen, Wen
    Zhang, Shaoting
    Li, Kang
    Lao, Qicheng
    IEEE TRANSACTIONS ON MEDICAL IMAGING, 2024, 43 (01) : 416 - 426
  • [27] Parallel Detection-and-Segmentation Learning for Weakly Supervised Instance
    Shen, Yunhang
    Cao, Liujuan
    Chen, Zhiwei
    Zhang, Baochang
    Su, Chi
    Wu, Yongjian
    Huang, Feiyue
    Ji, Rongrong
    2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021), 2021, : 8178 - 8188
  • [28] A deep learning approach for insulator instance segmentation and defect detection
    Antwi-Bekoe, Eldad
    Liu, Guisong
    Ainam, Jean-Paul
    Sun, Guolin
    Xie, Xiurui
    Neural Computing and Applications, 2022, 34 (09) : 7253 - 7269
  • [29] Visual Odometry Based on Feature Points of Instance Segmentation in Indoor Environment
    Kuang, Jia
    Cheng, Lei
    Cai, Zhihao
    Xiang, Honglian
    2020 CHINESE AUTOMATION CONGRESS (CAC 2020), 2020, : 2017 - 2023
  • [30] OWS-Seg: Online Weakly Supervised Video Instance Segmentation via Contrastive Learning
    Ning, Yuanxiang
    Li, Fei
    Dong, Mengping
    Li, Zhenbo
    ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING, ICANN 2023, PT VII, 2023, 14260 : 476 - 488