Career Profile

My undergraduate course on pattern recognition lit my passion for research. After I graduated with a Bachelor's degree in Intelligent Engineering (2011), I directly became a PhD candidate and obtained a doctoral degree from the school of artificial intelligence, Xidian University, China (2018). In 2019, I joined RR@NTU as a research scientist in Singapore. The half-business and half-academic working environments of RR@NTU helped me grow both professionally and personally. At the beginning of 2022, I started my new journey with SCRIPT@NTU. My current research interests include few-shot object detection for images/videos (FSOD), few-shot instance segmentation for images/videos (FSIS), 3D Steganography and optimization algorithms for power management systems (PMS).

Education

PH.D., School of Artificial Intelligence

2011 - 2018
Xidian University, China

Visiting Student

2013 - 2014
Leiden University, the Netherland

B.S. School of Electronic Engineering}{Xidian University

2007 - 2011
Xidian University, China

Publications

Adaptive multi-task learning for few-shot object detection
Ren, Yan and Li, Yanling and Kong, Adams Wai-Kin
European Conference on Computer Vision, 2024
Real-Time Shipboard Power Management Based on Monte-Carlo Tree Search
Ren, Yan and Kong, Adams Wai-Kin and Wang, Yi
IEEE Transactions on Power Systems, 2022
Power system design optimization for a ferry using hybrid-shaft generators
Oo, Thant Zin and Ren, Yan and Kong, Adams Wai-Kin and Wang, Yi and Liu, Xiong
IEEE Transactions on Power Systems, 2021
A survey on image and video cosegmentation: Methods, challenges and analyses
Ren, Yan and Kong, Adams Wai Kin and Jiao, Licheng
Pattern Recognition, 2020
Mutual Learning Cosegmentation via Tree Structured Sparsity and Tree Graph Matching
Yan Ren, Licheng Jiao, Shuyuan Yang, Shuang Wang
IEEE Transaction on Image Processing, 2018, DOI: 10.1109/TIP.2018.2842207.
A novel eye localization method with rotation invariance
Yan Ren, Shuang Wang, Biao Hou, Jingjing Ma
IEEE Transaction on Image Processing, 2014, 23(1):226-239.

Projects

Adaptive Multi-task Learning for Few-shot Object Detection

2022 - 2024

The majority of few-shot object detection methods use a shared feature map for both classification and localization, despite the conflicting requirements of these two tasks. Localization needs scale and positional sensitive features, whereas classification requires features that are robust to scale and positional variations. Although few methods have recognized this challenge and attempted to address it, they may not provide a comprehensive resolution to the issue. To overcome the contradictory preferences between classification and localization in few-shot object detection, an adaptive multi-task learning method, featuring a novel precision-driven gradient balancer, is proposed. This balancer effectively mitigates the conflicts by dynamically adjusting the backward gradient ratios for both tasks. Furthermore, a knowledge distillation and classification refinement scheme based on CLIP is introduced, aiming to enhance individual tasks by leveraging the capabilities of large vision-language models. Experimental results of the proposed method consistently show improvements over strong few-shot detection baselines on benchmark datasets.

MARINE POWER SYSTEM DESIGN AND ENERGY MANAGEMENT

2019 -- 2022

Her primary research focusses on few-shot learning, particularly in scenarios with limited annotated data. Many privacy-preserving applications encounter challenges due to the lack of sufficient data, such as privacy-sensitive images or objects in surveillance systems, or privacy-preserving object detection for medical image analysis. Few-shot learning shows promise in such cases, where the availability of labelled data is restricted, or privacy concerns prevent the collection and sharing of sensitive data.

WEAKLY SUPERVISED AND UNSUPERVISED COMMONALITY ANALYSIS

2017 - 2019

A first comprehensive survey covering both image and video cosegmentation is proposed, which are two newly emerged and under developed fields under the domain of image processing and video understanding

IMAGE COSEGMENTATION WITH WEIGHTED TREE STRUCTURED SPARSITY

2013 - 2017

Under weakly supervised condition, a novel unified mutual learning framework is proposed, which complement the adavantages of both tree-graph matching and structured sparsity to fully exploit the spatial information of hierarchical structures and accomplish the task of image cosegmentation.

ROTATION INVARIANT OBJECT DETECTION and LOCALIZATION

2011 - 2013

A Novel Rotation Invariant Learning Method for Object Detection and Localization via Sparse Representation

AUTOMATIC COSMETIC LENSES

2010 - 2011

This Program of Summer Internship is aim at automatically beautify human eyes in face images. The two main steps are the localization of eyes and the warp of eye patterns. The result of the algorithm looks like give cosmetic contacts to eyes

INTELLIGENT CLASSROOM BASED ON IMAGE UNDERSTANDING

2009 - 2010

National University Student Innovation Program with a budget of 12,000 RMB

SPARK CUP ACADEMIC COMPETITION

2008 - 2009

A single spark can start a prairie fire.This competition is my stepping-stone to the world of engineering.

Skills & Proficiency

Python

Pytorch

Matlab

LaTex

Linux