About Me

My name is Chengxin Liu. I am a Postdoctoral Researcher at Korea Advanced Institute of Science and Technology (KAIST), working with Prof. Tae-Hyun Oh. I received my PhD degree from Huazhong University of Science and Technology (HUST), advised by Prof. Zhiguo Cao and Prof. Hao Lu. Before that, I got my bachelor’s degree from HUST.

My research interests include computer vision and machine learning, with a particular emphasis on

• Multimodal learning (e.g., vision-language models),
• Data-efficient learning (e.g., learning with noisy/unlabeled data),
• Dense prediction (e.g., object detection, object counting, and segmentation),
• Generative models.

🔥 News

• 2026.02: Two papers accepted by CVPR 2026.
• 2026.02: Received AAAI’2026 Outstanding Program Committee Award.
• 2025.07: One paper accepted by ACM MM 2025.
• 2025.04: Received KAIST Jang Young Sil Fellowship.

📝 Publications [Full Paper]

Vision-Language Models

• CVPR 2026 Aligning What Vision-Language Models See and Perceive with Adaptive Information Flow

  Chengxin Liu, Wonseok Choi, Chenshuang Zhang, Tae-Hyun Oh

  IEEE/CVF Conference on Computer Vision and Pattern Recognition 2026 (CVPR 2026)

  [project page]

• CVPR 2026 SVHalluc: Benchmarking Speech–Vision Hallucination in Audio-Visual Large Language Models

  Chenshuang Zhang, Kyeong Seon Kim, Chengxin Liu, Tae-Hyun Oh

  IEEE/CVF Conference on Computer Vision and Pattern Recognition 2026 (CVPR 2026)

  [project page]

Object Detection

• ECCV 2022 Robust Object Detection With Inaccurate Bounding Boxes

  Chengxin Liu, Kewei Wang, Hao Lu, Zhiguo Cao, Ziming Zhang

  European Conference on Computer Vision 2022 (ECCV 2022)

  [paper] [code]

• Plant Phenomics Dynamic Color Transform Networks for Wheat Head Detection

  Chengxin Liu, Kewei Wang, Hao Lu, Zhiguo Cao

  Plant Phenomics, 2022

  [paper] [code]

• ICCVW 2021 Dynamic Color Transform for Wheat Head Detection

  Chengxin Liu, Kewei Wang, Hao Lu, Zhiguo Cao

  IEEE/CVF International Conference on Computer Vision Workshops 2021 (ICCVW 2021)

  [paper] [code]

• TGRS Interior Attention-Aware Network for Infrared Small Target Detection

  Kewei Wang, Shuaiyuan Du, Chengxin Liu, Zhiguo Cao

  IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2022

  [paper] [code]

• IGARSS CLIP-guided Source-free Object Detection in Aerial Images

  Nanqing Liu, Xun Xu, Yongyi Su, Chengxin Liu, Peiliang Gong, Heng-Chao Li

  IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 2024

  [paper] [code]

Object Counting

• ICCV 2023 Point-Query Quadtree for Crowd Counting, Localization, and More

  Chengxin Liu, Hao Lu, Zhiguo Cao, Tongliang Liu

  IEEE/CVF International Conference on Computer Vision 2023 (ICCV 2023)

  [paper] [code]

• IJCV From Open Set to Closed Set: Supervised Spatial Divide-and-Conquer for Object Counting

  Haipeng Xiong, Hao Lu, Chengxin Liu, Liang Liu, Chunhua Shen, Zhiguo Cao

  International Journal of Computer Vision (IJCV), 2023

  [paper] [code]

• ICCV 2019 From Open Set to Closed Set: Counting Objects by Spatial Divide-and-Conquer

  Haipeng Xiong, Hao Lu, Chengxin Liu, Liang Liu, Chunhua Shen, Zhiguo Cao

  IEEE/CVF International Conference on Computer Vision 2019 (ICCV 2019)

  [paper] [code]

• CVPR 2022 Represent, Compare, and Learn: A Similarity-Aware Framework for Class-Agnostic Counting

  Min Shi, Hao Lu, Chen Feng, Chengxin Liu, Zhiguo Cao

  IEEE/CVF Conference on Computer Vision and Pattern Recognition 2022 (CVPR 2022)

  [paper] [code]

Segmentation and Tracking

• GRSL Multispectral Semantic Land Cover Segmentation From Aerial Imagery With Deep Encoder–Decoder Network

  Chengxin Liu, Shuaiyuan Du, Hao Lu, Dehui Li, Zhiguo Cao

  IEEE Geoscience and Remote Sensing Letters (GRSL), 2020

  [paper]

• ICPR 2020 Exploiting Distilled Learning for Deep Siamese Tracking

  Chengxin Liu, Zhiguo Cao, Wei Li, Yang Xiao, Shuaiyuan Du, Angfan Zhu

  International Conference on Pattern Recognition 2020 (ICPR 2020)

  [paper]

Anime Talking Head

• TCSVT Hierarchical Feature Warping and Blending for Talking Head Animation.

  Jiale Zhang, Chengxin Liu, Ke Xian, Zhiguo Cao

  IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2024

  [paper]

• PR Large Motion Anime Head Animation Using a Cascade Pose Transform Network

  Jiale Zhang, Chengxin Liu, Ke Xian, Zhiguo Cao

  Pattern Recognition (PR), 2023

  [paper] [code]

• ACCV 2020 CPTNet: Cascade Pose Transform Network for Single Image Talking Head Animation

  Jiale Zhang, Ke Xian, Chengxin Liu, Yinpeng Chen, Zhiguo Cao, Weicai Zhong

  Asian Conference on Computer Vision 2020 (ACCV 2020)

  [paper]

Experience

• Visiting PhD Student @ Institute for Infocomm Research (I2R), A*STAR, Singapore

  Time: 2023.11 - 2024.08, Mentor: Dr. Xun Xu

Academic Service

• Conference Reviewer: CVPR, ICCV, ECCV, NeurIPS, ICLR, ICML, AAAI, ACM MM, WACV, ACCV, PRCV
• Journal Reviewer: TPAMI, IJCV, TIP, TMM, Plant Phenomics

Competition

• Second Prize in China Graduate AI Innovation Competition (10000 RMB bonus), 2022
• 2nd Place in Global Wheat Challenge 2021 @ ICCV (1000 USD bonus), 2021