Guosheng Lin

School of Computer Science,
The University of Adelaide, SA 5005.
Email: guosheng.lin{At}adelaide.edu.au

ARC Center of Excellence for Robotic Vision (ACRV) (http://roboticvision.org/)
Australian Centre for Visual Technologies (ACVT) (http://www.acvt.com.au/)

I am a research fellow at ACRV and ACVT  working with Prof. Chunhua shen, Prof. Ian Reid.
My research interests are in machine learning and computer vision applications, particularly in visual recognition problems.
I completed my PhD degree at the University of Adelaide (2010-2014), supervised by Prof. Chunhua Shen (principal),  Prof. David Suter and Dr.Tat-Jun Chin.
I received my bachelor and master degree from South China University of Technology (SCUT) in 2007 and 2010 respectively.

Google scholar page: [link]

Awards
:

Google PhD fellowship, Australia, 2014, see here or here.

Teaching:

Working papers:


Conference publications:

--------------------------------------------------------------------------------------------------------------------------------------------------------------
  •  "RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentation";
Guosheng Lin, Anton Milan, Chunhua Shen, Ian Reid; IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2017; [arXiv]

Top performance on PASCAL VOC 2012 dataset (Links: Leaderboard),
**NEW**  our source code and trained models are available at: https://github.com/guosheng/refinenet
(MATLAB based framework for semantic segmentation and dense preidction)
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  •  "Sequential Person Recognition in Photo Albums with a Recurrent Network";
Yao Li, Guosheng Lin, Bohan Zhuang, Lingqiao Liu, Chunhua Shen, Anton van den Hengel; IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2017; [arXiv]
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • Learning Multi-level Region Consistency with Dense Multi-label Networks for Semantic Segmentation
Tong Shen*, Guosheng Lin*, Chunhua Shen, Ian Reid; International Joint Conference on Artificial Intelligence (IJCAI) 2017; [arXiv]
( * indicates equal contribution );
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Efficient Piecewise Training of Deep Structured Models for Semantic Segmentation"
Guosheng Lin,  Chunhua Shen, Anton van den Hengel, Ian Reid; IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, Spotlight presentation; [arXiv]

Links: PASCAL VOC 2012 dataset Leaderboard, Cityscapes dataset Leaderboard.
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Fast training of triplet-based deep binary embedding networks "
Bohan Zhuang, Guosheng Lin,  Chunhua Shen, Ian Reid; IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016; [arXiv]
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Deeply Learning the Messages in Message Passing Inference"
Guosheng Lin,  Chunhua Shen, Ian Reid, Anton van den Hengel;  Advances in Neural Information Processing Systems (NIPS), 2015, [arXiv]
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Deep Convolutional Neural Fields for Depth Estimation from a Single Image"
Fayao Liu, Chunhua Shen, Guosheng Lin; IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015; [arXiv]

--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Sequence Searching With Deep-Learnt Depth for Condition and Viewpoint-Invariant Route-Based Place Recognition"
Michael Milford, Chunhua Shen, Stephanie Lowry, Niko Suenderhauf, Sareh Shirazi, Guosheng Lin, Fayao Liu,
Edward Pepperell, Cesar Lerma, Ben Upcroft, Ian Reid; The 6th international Workshop on Computer Vision in Vehicle Technology, CVPR2015 workshop.
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Fast Supervised Hashing with Decision Trees for High-Dimensional Data";
Guosheng Lin, Chunhua Shen, Qinfeng Shi, Anton van den Hengel and David Suter; IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2014; [PDF]

Code: https://bitbucket.org/chhshen/fasthash/
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Optimizing Ranking Measures For Compact Binary Code Learning"
Guosheng Lin,  Chunhua Shen, Jianxin Wu; European Conference on Computer Vision (ECCV), 2014; [PDF]

Extensions of this method for efficient training are also included
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "A General Two-Step Approach to Learning-Based Hashing";
Guosheng Lin, Chunhua Shen, David Suter and Anton van den Hengel; International Conference on Computer Vision  (ICCV), 2013; [PDF] or [PDF]

--------------------------------------------------------------------------------------------------------------------------------------------------------------
  •  "Learning Hash Functions Using Column Generation";
Xi Li*, Guosheng Lin*, Chunhua Shen, Anton van den Hengel and Anthony Dick; International Conference on Machine Learning (ICML), 2013; Oral Presentation; [PDF]
 ( * indicates equal contribution );

--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Approximate Constraint Generation for Efficient Structured Boosting";
Guosheng Lin, Chunhua Shen and Anton van den Hengel; International Conference on Image Processing (ICIP), 2013; [PDF] or [PDF]
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Fast Training of Effective Multi-class Boosting Using Coordinate Descent Optimization";
Guosheng Lin, Chunhua Shen, Anton van den Hengel and David Suter;  Asian Conference on Computer Vision (ACCV), 2012; [PDF] or [PDF]

Code: https://bitbucket.org/guosheng/fast-multiboost-cw
--------------------------------------------------------------------------------------------------------------------------------------------------------------





Journal publications:


--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Exploring Context with Deep Structured models for Semantic Segmentation";
Guosheng Lin, Chunhua Shen, Anton van den Hengel, Ian Reid;  IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2017; [arXiv]
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Supervised Hashing Using Graph Cuts and Boosted Decision Trees";
Guosheng Lin, Chunhua Shen, Anton van den Hengel;  IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2015; [PDF] or [arXiv]

Code: https://bitbucket.org/chhshen/fasthash/
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Structured Learning of Binary Codes with Column Generation"
Guosheng Lin, Fayao Liu, Chunhua Shen, Jianxin Wu, Heng Tao Shen; International Journal of Computer Vision (IJCV); 2017;

Code: https://bitbucket.org/guosheng/structhash
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "Learning Depth from Single Monocular Images Using Deep Convolutional Neural Fields"
Fayao Liu, Chunhua Shen, Guosheng Lin, Ian Reid; IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI); 2016; [arXiv] or [PDF]

Code: https://bitbucket.org/fayao/dcnf-fcsp/
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • "StructBoost: Boosting Methods For Predicting Structured Output Variables";
Chunhua Shen, Guosheng Lin, Anton van den Hengel; IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2014; [PDF] or [arXiv]

The following code provides the implementation of cutting-plane and an example of column generation:
https://bitbucket.org/guosheng/structhash
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • Discriminative Training of Deep Fully-connected Continuous CRF with Task-specific Loss
Fayao Liu, Guosheng Lin, Chunhua Shen; IEEE Transactions on Image Processing (TIP), 2017; [PDF]
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • Structured Learning of Tree Potentials in CRF for Image Segmentation
Fayao Liu, Guosheng Lin, Ruizhi Qiao, Chunhua Shen; IEEE Transactions on Neural Networks and Learning Systems (TNNLS), 2017;
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  •  "CRF Learning with CNN Features for Image Segmentation";
Fayao Liu, Guosheng Lin, Chunhua Shen; Pattern Recognition (PR), 2015; [PDF]
--------------------------------------------------------------------------------------------------------------------------------------------------------------
  • Crowd Counting via Weighted VLAD on Dense Attribute Feature Maps
Biyun Sheng, Chunhua Shen, Guosheng Lin, Jun Li, Wankou Yang, Changyin Sun; IEEE Transactions on Circuits and Systems for Video Technology (T-CSVT); 2016; [PDF]
--------------------------------------------------------------------------------------------------------------------------------------------------------------