Ge-Peng (Daniel) Ji

Ge-Peng (Daniel) Ji

Ph.D. Student (CECC@ANU)

About Me

In the fall of 2022, I began my Ph.D. studies at the College of Engineering, Computing & Cybernetics (CECC) at the Australian National University. My supervisor is Prof. Nick Barnes, and I also have the privilege of receiving academic guidance from Prof. Deng-Ping Fan. Prior to this, in 2021, I obtained my Master’s degree from School of Computer Science, Wuhan University.

Research Interests

My primary research interest is to solve complex perception problems using computer vision and machine learning techniques, particularly by developing efficient and scalable algorithms for dense prediction tasks in real-world scenarios. These tasks include but are not limited to:

  • Complex Scene Understanding: Camouflaged/Salient Object Discovery on different data modalities (e.g., RGB, RGB-Depth, RGB-Thermal, Light-Field)
  • Medical Image Segmentation: Colorectal Polyp Segmentation, Lung Infection Segmentation
  • Video Analysis: Video Object Segmentation, Video Camouflaged/Salient Object Detection, Motion Segmentation
  • Open-World Scene Understanding: Out-of-distribution Detection, Open-vocabulary, Large Multimodal Models (LMMs)

Recent Updates

  • [Sep 28, 2023] We conduct empirical studies on the Segment Anything Model (SAM) by investigating its abilty on three concealed scenarios. (arXiv & 公众号解读), which was accepted by SCIS journal.
  • [Apr 25, 2023] We release a comprehensive survey paper about ‘Deep Concealed Scene Understanding’. (arXiv & Project & 公众号解读)
  • [Feb 14, 2023] Call for paper: Multi-Modal Representation Learning, Special Issue in Machine Intelligence Research (MIR) journal, 2023 (Submission Deadline: July 1, 2023)
  • [Nov 26, 2022] We release two awesome paper reading lists for the camouflaged object detection (COD) and video polyp segmentation (VPS)
  • [Nov 04, 2022] One paper about video polyp segmentation is published in Machine Intelligence Research (MIR) journal. (Open Access - Springer & MIR News)
  • [Oct 18, 2022] Our paper about video object segmentation is published in Computational Visual Media (CVM) journal. (Open Access)
  • [Aug 06, 2022] One paper about efficient camouflaged object detection was accepted by Machine Intelligence Research (MIR) journal. (Open Access - Springer)

Viewing Stats

Contact

Email: echo Z2VwZW5nYWkuamlAZ21haWwuY29tCg== | base64 --decode