Ching-Yu Tseng
email: sty61010 [at] gmail.com

Ching-Yu is a graduate research assistant at Communication & Multimedia Lab at National Taiwan University (NTU), supervised by Prof. Winston H. Hsu. Previously, he was an undergraduate research assistant working with Prof. Shan-Hung Wu in the Data Lab at National Tsing Hua University (NTHU). His research focuses on Machine Learning, Computer Vision, and Autonomous Driving. He has won the top 3rd in the top conference NeurIPS2021 competition. Recently, he has been working on camera-based perception for autonomous driving. His work has been accepted by The 40th IEEE Conference on Robotics and Automation (ICRA 2023). Ching-Yu is a student member of IEEE and RAS.

Ching-Yu received his Bachelor's degree in Computer Science from National Tsing Hua University (NTHU) in 2021, with an overall GPA of: 4.02/4.3 and a class rank top 11%. Ching-Yu was honored with the NTHU Dean’s List Award for the top 5% of students twice in 2017-2021, He also received his Master's degree in Computer Science and Information Engineering from National Taiwan University (NTU) in 2023, with an overall GPA: 4.08/4.3. If you would like to learn more about me, please see my [CV] or contact me at sty61010 [at] gmail.com.

sym

NTU
M.Sc. in CSIE
Sep. 21 - Present

sym

NTHU
B.Sc. in CS
Sep. 17 - Jun. 21

sym

Shanghai Jiao Tong University
Visiting Researcher
Jul. 19 - Sep. 19

  Publications
sym

Revisiting Depth-guided Methods for Monocular 3D Object Detection by Hierarchical Balanced Depth
Yi-Rong Chen, Ching-Yu Tseng, Yi-Syuan Liou, Tsung-Han Wu, Winston H. Hsu
2023 Conference on Robot Learning (CoRL 2023)
Under Submission

Optimized Monocular 3D Object Detection with the depth-guided method by leveraging RoIAlign to capture both scene-level depth relationships and object-specific depth characteristics.

sym

CrossDTR: Cross-view and Depth-guided Transformers for 3D Object Detection
Ching-Yu Tseng, Yi-Rong Chen, Hsin-Ying Lee, Tsung-Han Wu, Wen-Chin Chen, Winston H. Hsu
The 40th IEEE Conference on Robotics and Automation (ICRA 2023)
Paper / Video / Code / Project

Proposed a Multi-view 3D Object Detection framework with depth guidance.

sym

3rd Place Solution for NeurIPS 2021 Shifts Challenge: Vehicle Motion Prediction
Ching-Yu Tseng, Po-Shao Lin, Yu-Jia Liou, Kuan-Chih Huang, Winston H. Hsu
The 35th Conference on Neural Information Processing Systems (NeurIPS 2021) Competition
Paper / Video / Competition

Proposed a novel Motion Prediction framework with a self-attention mechanism among different timestamps.

sym

Learning from 2D: Contrastive Pixel-to-Point Knowledge Transfer for 3D Pretraining
Yueh-Cheng Liu, Yu-Kai Huang, Hung-Yueh Chiang, Hung-Ting Su, Zhe-Yu Liu, Chin-Tang Chen, Ching-Yu Tseng, Winston H. Hsu
ArXiv preprint, 2021; Citation: 33
Paper

Proposed a novel 3D pretraining method by leveraging 2D networks learned from rich 2D datasets.

  Side Projects
sym

Survey and Implementation of Facial Detection and Convolution Neural Network [OpenCV, Raspberry Pi, Keras]
Final Project when I was a visiting researcher in SJTU
Video / Code

  • Designed an application supporting real‑time face recognition. Deployed on the webcam of a laptop and Raspberry Pi.
  • Utilized OpenCV for face detection and CNN architecture (i.e., FaceNet) for face feature extraction.
  • Trained a Cascade Classifier on extracted facial features on the self‑collected facial dataset
  • sym

    URcode [Unity, C#, Blender]
    Honorable Mentioned Project of ”Game Programming”
    Video / Code

  • Led 4 people to develop an immersive game with the theme of QRcode in Unity.
  • Design Interact between player and environment.
  • Cooperate with members from the Department of Art at NTHU.
  • sym

    Arduino Remote Control Tank [C++, Arduino, Embedding System]
    Outstanding Final Project of ”Embedding System”
    Video / Code

  • Use a PS2 controller to control the self-develop tank with Bluetooth.
  • Use an ultrasonic sensor to realize random autonomous driving.
  • Combine applications from multi-sensor (Bluetooth, ultrasonic, light).
  •   Teaching
      Honors & Awards
    • 3rd Place, NeurIPS 2021 Competition, 2021 (NeurIPS 2021 Shifts Challenge: Vehicle Motion Prediction)

    • College Student Research Scholarship, Ministry of Science and Technology, 2020 (Research scholarship from Ministry of Science and Technology.)

    • 1st place in DataLab Cup, Kaggle Competition of CS5656 Deep Learning, NTHU, 2020 (Deep Learning Competition in Department of Computer Science, National Tsing Hua University)

    • Summer Oversea Experience Scholarship, NTHU, 2018 (Exchange program to Shanghai Jiao Tong University for visiting researcher.)

    • Academic Achievement Award * 2 times, NTHU, Fall 2018, Spring 2021 (Top 5% GPA in each semester)