Co-saliency based visual object co-segmentation and co-localization
Koteswar Rao Jerripothula · 2017
Automatic foreground segmentation and localization in images or videos are very important and basic problems in computer vision.Due to lacking of sufficient information about the foreground object in a single image or a video, these tasks usually become very difficult.However, if a set of similar images (where foreground objects are of the same category) are provided for joint processing, the job becomes little easier because we can exploit the available commonness clue.Thus, by jointly processing similar images together we provide a kind of weak supervision to the system.Such a task of segmenting out the common foreground visual objects through joint processing of similar images is known as co-segmentation.Similarly, the task of localizing (proving bounding box to) the common foreground visual objects is known as colocalization.Co-segmentation and co-localization tasks have applications in image retrieval, image synthesis, datasets generation, object recognition, video surveillance, action recognition, etc.However, such joint processing brings in new challenges to handle: (i) variation in terms When I first came to NTU, I hardly had any prior research experience except the FYP that I did during my undergraduate studies.I feel extremely grateful towards my supervisors, Prof. Jianfei Cai and Prof. Junsong Yuan, for giving me the opportunity to pursue my PhD under their expert guidance, which helped me a lot to develop gradually and progressively as a researcher.Their timely and precise inputs taught me many important lessons both in the research and my personal life which I will always relish and feel inspired by.I would also like to thank Dr. Lu Jiangbo for his energetic inputs, which inspired me further.I thank my teacher for giving me the right motivation behind why I should do my PhD.His golden words about PhD, "It's 99% perspiration and only 1% inspiration, despite that what drives it is right motivation", sank into my heart and helped me overcome my struggles during the PhD.Lastly, I would like to thank my parents, mentors and friends for giving me all the encouragement and support that I needed for successfully completing this thesis.