CS Events Monthly View
PhD DefenseTowards Active and Interactive Visual Learning |
|
||
Friday, September 29, 2017, 11:00am |
|||
Modern computer vision models mostly rely on massive human annotated dataset for supervised training. Models are typically learned from the supervision of a static dataset with a passive manner.
This work explores three new settings when huge dataset supervisions are scarce and novel learning paradigms beyond passive training are proposed. First, we addressed the active learning for histopathological images diagonal systems, and proposed an active selection algorithm via constrained submodular function maximization, the result shows the active selected training set is compact and outperform state of the art selection algorithms. Second, we proposed a novel semantic amodal segmentation task in which occluded object segmentation masks are predicted, and synthetic hard occluded examples are actively generated for training. Third, we addressed learning a visual grounding task via natural language interactions, in which two agents are trained to interact via interpretable dialogue to achieve a common goal.
Speaker: Yan Zhu
Bio
NULL
Location : CBIM 22
Committee:
Prof. Dimitris Metaxas (Chair), Prof. Kostas Bekris, Prof. Konstantinos Michmizos, Prof. Dimitris Samaras (Stony Brook University)
Event Type: PhD Defense
:
Organization:
Dept. of Computer Science