Currently, I am a research intern at CMU Robotics Institute, supervised by Prof.Jun-Yan Zhu. I am broadly interested in 3D vision and robot learning, especially realistic 3D content generation using physical and neural representations.
.
We generate images of the 3D objects taken from 3D shape repositories (e.g., ShapeNet and Objaverse), render them from a variety of poses and viewing directions, compute the edge maps of the rendered images, and use these edge maps as visual prompts to generate realistic images.
Left-right Discrepancy for Adversarial Attack on Stereo Networks Pengfei Wang, Xiaofei Hui, Beijia Lu, Nimrod Lilith, Jun Liu, Sameer Alam
Submitted to CVPR, 2024 arXiv
We introduce an adversarial attack method that generates perturbations to amplify discrepancies between left and right image features in stereo matching neural networks.
Dual-View Selective Instance Segmentation Network for Unstained Live Adherent Cells in Differential Interference Contrast Images
Fei Pan, Yutong Wu, Kangning Cui, Shuxun Chen, Yanfang Li, Yaofang Liu, Adnan Shakoor, Han Zhao, Beijia Lu, Shaohua Zhi, Raymond Chan, Dong Sun
Submitted to Medical Image Anaylsis arXiv
We develop a novel deep-learning algorithm for segmenting unstained adherent cells in DIC images, achieving significant advancements in cell instance segmentation accuracy.