student profile: Mr Kaelem Yan


Map

Thesis work

Thesis title: Recognition and Segmentation of Targeting Objects from Multimodality Imaging Data

Supervisors: Xiu WANG , David FENG

Thesis abstract:

Object detection and segmentation, which discover a set of regions containing target object instances, is an important task in computer vision. Most conventional methods of object detection and segmentation rely on local features derived from image cues for object detection and segmentation. Recently, instead of the carefully designed local features, some investigators employ deep neural network (DNN) to extract global features for detection and segmentation. Though global features from DNN are more powerful to represent the intrinsic and topological structures, only depending on the single level feature may lead to missing local details during the data feed-forward in DNN. Therefore, we explore the optimal combination of multi-level features and consequently the results of object detection and segmentation could be rewarded by optimizing the single level features. High-level features are more powerful to represent the intrinsic and semantic structures of data, but usually lose some local details during the data feedforward layer by layer. On the contrary, as low-level features are usually extracted within a small atomic homogenous region (such as image patch and superpixel), they can intuitively depict the local appearance and tiny change. Hence, we hypothesize that an optimal combination of these multi-level features could improve the current object recognition and segmentation results.

Selected publications

Download citations: PDF RTF Endnote

Journals

  • Xu, Y., Yan, K., Kim, J., Wang, X., Li, C., Su, L., Yu, S., Xu, X., Feng, D. (2017). Dual-stage deep learning framework for pigment epithelium detachment segmentation in polypoidal choroidal vasculopathy. Biomedical Optics Express, 8(9), 4061-4076. [More Information]

Conferences

  • Yan, K., Zheng, C., Huang, Q., Kim, J., Feng, D., Wang, X. (2018). Prior Knowledge Driven Energy for Saliency Detection. 2018 15th International Conference on Control, Automation, Robotics and Vision (ICARCV), Singapore: Institute of Electrical and Electronics Engineers (IEEE). [More Information]
  • Yan, K., Li, C., Wang, X., Li, A., Yuan, Y., Kim, J., Feng, D. (2016). Adaptive background search and foreground estimation for saliency detection via comprehensive autoencoder. 23rd IEEE International Conference on Image Processing (ICIP 2016), Piscataway: Institute of Electrical and Electronics Engineers (IEEE). [More Information]
  • Yan, K., Li, C., Wang, X., Li, A., Yuan, Y., Feng, D., Khadra, M., Kim, J. (2016). Automatic prostate segmentation on MR images with deep network and graph model. 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2016), Orlando: Institute of Electrical and Electronics Engineers (IEEE). [More Information]
  • Yan, K., Li, C., Wang, X., Yuan, Y., Li, A., Kim, J., Li, B., Feng, D. (2016). Comprehensive autoencoder for prostate recognition on MR images. 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI): From Nano to Macro, Piscataway: Institute of Electrical and Electronics Engineers (IEEE). [More Information]
  • Li, A., Wang, X., Yan, K., Li, C., Feng, D. (2016). Multilevel affinity graph for unsupervised image segmentation. 23rd IEEE International Conference on Image Processing (ICIP 2016), Piscataway: Institute of Electrical and Electronics Engineers (IEEE). [More Information]

2018

  • Yan, K., Zheng, C., Huang, Q., Kim, J., Feng, D., Wang, X. (2018). Prior Knowledge Driven Energy for Saliency Detection. 2018 15th International Conference on Control, Automation, Robotics and Vision (ICARCV), Singapore: Institute of Electrical and Electronics Engineers (IEEE). [More Information]

2017

  • Xu, Y., Yan, K., Kim, J., Wang, X., Li, C., Su, L., Yu, S., Xu, X., Feng, D. (2017). Dual-stage deep learning framework for pigment epithelium detachment segmentation in polypoidal choroidal vasculopathy. Biomedical Optics Express, 8(9), 4061-4076. [More Information]

2016

  • Yan, K., Li, C., Wang, X., Li, A., Yuan, Y., Kim, J., Feng, D. (2016). Adaptive background search and foreground estimation for saliency detection via comprehensive autoencoder. 23rd IEEE International Conference on Image Processing (ICIP 2016), Piscataway: Institute of Electrical and Electronics Engineers (IEEE). [More Information]
  • Yan, K., Li, C., Wang, X., Li, A., Yuan, Y., Feng, D., Khadra, M., Kim, J. (2016). Automatic prostate segmentation on MR images with deep network and graph model. 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2016), Orlando: Institute of Electrical and Electronics Engineers (IEEE). [More Information]
  • Yan, K., Li, C., Wang, X., Yuan, Y., Li, A., Kim, J., Li, B., Feng, D. (2016). Comprehensive autoencoder for prostate recognition on MR images. 2016 IEEE 13th International Symposium on Biomedical Imaging (ISBI): From Nano to Macro, Piscataway: Institute of Electrical and Electronics Engineers (IEEE). [More Information]
  • Li, A., Wang, X., Yan, K., Li, C., Feng, D. (2016). Multilevel affinity graph for unsupervised image segmentation. 23rd IEEE International Conference on Image Processing (ICIP 2016), Piscataway: Institute of Electrical and Electronics Engineers (IEEE). [More Information]

Note: This profile is for a student at the University of Sydney. Views presented here are not necessarily those of the University.