Chunlai Wang


ehem. Wiss. Mitarbeiter

E-Mail               Link

Adresse            Universität Stuttgart
                         Pfaffenwaldring 47
                         70550 Stuttgart


My research interests are in the areas of computer vision and machine learning. In particular, I'm interested in object-level image segmentation, e.g. salient object detection and segmentation, semantic image segmentation, instance-level image segmentation. Techniques to solve these problems include both unsupervised and supervised methods.

Student thesis

Fusing Additional Information for Object Class Extension in Semantic Image Segmentation Using Convolutional Neural Network (ongoing)

Context-Insensitive Convolutional Neural Network for Semantic Segmentation with Refined Object Boundaries (ongoing)

Exploring Contextual Information in Deep CNN for Semantic Image Segmentation (ongoing)

Unsupervised Object-Level Image Segmentation using CNN with Spatial Pyramid Pooling (ongoing)

Instance-Aware Semantic Image Segmentation Using Deep Neural Networks (ongoing)

Unsupervised Image Segmentation Using Features Learned from Convolutional Network (completed)

Salient Object Segmentation Using Deep Convolutional Neural Network (completed)

Segmentierung von Objektkandidaten auf der Straße (completed)

Convolutional Neural Network for Low-Level Image Segmentation (completed)

Implementation, evaluation and enhancement of a fast "optical flow estimation" method (completed, extern)

Training Convolutional Neural Network for Semantic Segmentation in Consideration of Easy Extention for New Object Classes (completed)

Delving into Region Features for Single Image Foreground/Background Segmentation (completed)

Effectiveness Analysis of Using Saliency for Foreground Object Segmentation (completed)

Farb- und Formbasierte Segmentierung zur Detektion von Geschwindigkeitsbegrenzungszeichen unter Variation der Lichtbedingungen (completed)

Region segmentation for road detection (completed)

Unsupervised GMM Methods for Color Image Segmentation (completed)


Chunlai Wang, Bin Yang, Yiwen Liao 
Unsupervised Image Segmentation Using Convolutional Autoencoder with Total Variation Regularization as Preprocessing 
Proceedings of the 42th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP),March 2017, New Orleans, USA.

Chunlai Wang, Lukas Mauch, Ze Guo and Bin Yang 
On Semantic Image Segmentation Using Deep Convolutional Neural Network with Shortcuts and Easy Class Extension 
Proceedings of The Sixth International Conference on Image Processing Theory, Tools and Applications (IPTA),Dec. 2016, Oulu, Finland.

Chunlai Wang and Bin Yang 
Saliency-Guided Object Proposal for Refined Salient Region Detection 
Proceedings of the 2016 IEEE International Conference on Visual Communications and Image Processing (VCIP),Nov. 2016, Chengdu, China.

Chunlai Wang and Bin Yang 
An Unsupvervised Object-Level Image Segmentation Method Based on Foreground and Background Priors
Proceedings of the 2016 IEEE Southwest Symposium on Image Analysis and Interpretation (SSIAI),  March 2016, Santa Fe, New Mexico, USA.

Zum Seitenanfang