IIS Computer Vision & Machine Learning Lab

** IIS - CVML Update **


We are currently recruiting full-time research assistants and postdocs. Applicants that have background in computer vision and machine learning are welcome to apply for the positions.

Applications and inquiries can be emailed to liutyng@iis.sinica.edu.tw (Tyng-Luh Liu)




** Our paper entitled "Self-Supervised Sparse Representation for Video Anomaly Detection`` is accepted by ECCV 2022.


** Our paper entitled "Decoupled Contrastive Learning` is accepted by ECCV 2022.


** Our paper entitled "Capturing Humans in Motion: Temporal-Attentive 3D Human Pose and Shape Estimation from Monocular Video`` is accepted by CVPR 2022.


** Our paper entitled "SAGA: Self-Augmentation with Guided Attention for Representation Learning`` is accepted by ICASSP 2022.


** Our paper entitled “Pose Adaptive Dual Mixup for Few-Shot Single-View Reconstruction`` is accepted by AAAI 2022.


** Our paper entitled “Learning Unsupervised Metaformer for Anomaly Detection`` is accepted by ICCV 2021.


** Our paper entitled “Adaptive Image Transformer for One-Shot Object Detection” is accepted by CVPR 2021.


** Our paper entitled “Referring Image Segmentation via Language-Driven Attention`` is accepted by ICRA 2021.


** Our paper entitled “Adaptive and Generative Zero-Shot Learning” is accepted by ICLR 2021.


** Our paper entitled “Text-guided Graph Neural Networks for Referring 3D Instance Segmentation`` is accepted by AAAI 2021.


** Our paper entitled “Pose Ad` is accepted by ACM MM 2020.


** Our paper entitled “Self-similarity Student for Partial Label Histopathology Image Segmentation” is accepted by ECCV 2020.


** Our paper entitled “Query-driven Multi-Instance Learning”  is accepted by AAAI 2020.


** Our paper entitled “One-Shot Object Detection with Co-Attention and Co-Excitation” is accepted by NeurIPS 2019.


** Our paper entitled ``See-through-Text Grouping for Referring Image Segmentation`` is accepted by ICCV 2019.


** Our paper entitled ``Tell Me Where It is Still Blurry: Adversarial Blurred Region Mining and Refining`` is accepted by ACM MM 2019.


** Our paper entitled ``Unsupervised Meta-learning of Figure-Ground Segmentation via Imitating Visual Effects`` is accepted by AAAI 2019.


** Our paper entitled ``A2A: Attention to Attention Reasoning for Movie Question Answering`` is accepted by ACCV 2018.


** The joint effort with Prof. Hwann-Tzong Chen and his students (Shou-Yao Roy Tseng & Shao-Heng Tai)  has won the first place for Instance Segmentation Challenge in Robust Vision Challenge at CVPR 2018.

** The joint effort with Prof. Hwann-Tzong Chen and his students is the current co-leader on MovieQA  leader board. (06/01/2018)


** The joint work on classifying shot types in concert videos with IIS & CITI colleagues is accepted for IEEE TMM publication.


** The joint work on 360 saliency with Prof. Min Sun and his students (Hsien-Tzu Cheng et al.) at NTHU EE is accepted by CVPR 2018.

** Welcome Dr. Ding-Jie Chen to join the IIS-CVL lab.

** IIS-CVL lab has received a four-year  MOST grant on AI research. The title of the project is ``GAN-Inspired Deep Learning Techniques and Network Compression for Computer Vision Applications.``