Paper ID | IVMSP-29.5 | ||
Paper Title | NLKD: using coarse annotations for semantic segmentation based on knowledge distillation | ||
Authors | Dong Liang, Yun Du, College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, China; Han Sun, Liyan Zhang, Ningzhong Liu, Mingqiang Wei, Nanjing University of Aeronautics and Astronautics, China | ||
Session | IVMSP-29: Semantic Segmentation | ||
Location | Gather.Town | ||
Session Time: | Friday, 11 June, 13:00 - 13:45 | ||
Presentation Time: | Friday, 11 June, 13:00 - 13:45 | ||
Presentation | Poster | ||
Topic | Image, Video, and Multidimensional Signal Processing: [IVARS] Image & Video Analysis, Synthesis, and Retrieval | ||
IEEE Xplore Open Preview | Click here to view in IEEE Xplore | ||
Abstract | Modern supervised learning relies on a large amount of training data, yet there are many noisy annotations in real datasets. For semantic segmentation tasks, pixel-level annotation noise is typically located at the edge of an object, while pixels within objects are fine-annotated. We argue the coarse annotations can provide instructive supervised information to guide model training rather than be discarded. This paper proposes a noise learning framework based on knowledge distillation NLKD, to improve segmentation performance on unclean data. It utilizes a teacher network to guide the student network that constitutes the knowledge distillation process. The teacher and student generate the pseudo-labels and jointly evaluate the quality of annotations to generate weights for each sample. Experiments demonstrate the effectiveness of NLKD, and we observe better performance with boundary-aware teacher networks and evaluation metrics. Furthermore, the proposed approach is model-independent and easy to implement, appropriate for integration with other tasks and models. |