Scale-Hybrid Group Distillation with Knowledge Disentangling for Continual Semantic Segmentation
Author:
Song Zichen1ORCID, Zhang Xiaoliang1ORCID, Shi Zhaofeng1ORCID
Affiliation:
1. School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China
Abstract
Continual semantic segmentation (CSS) aims to learn new tasks sequentially and extract object(s) and stuff represented by pixel-level maps of new categories while preserving the original segmentation capabilities even when the old class data is absent. Current CSS methods typically preserve the capacities of segmenting old classes via knowledge distillation, which encounters the limitations of insufficient utilization of the semantic knowledge, i.e., only distilling the last layer of the feature encoder, and the semantic shift of background caused by directly distilling the entire feature map of the decoder. In this paper, we propose a novel CCS method based on scale-hybrid distillation and knowledge disentangling to address these limitations. Firstly, we propose a scale-hybrid group semantic distillation (SGD) method for encoding, which transfers the multi-scale knowledge from the old model’s feature encoder with group pooling refinement to improve the stability of new models. Then, the knowledge disentangling distillation (KDD) method for decoding is proposed to distillate feature maps with the guidance of the old class regions and reduce incorrect guides from old models towards better plasticity. Extensive experiments are conducted on the Pascal VOC and ADE20K datasets. Competitive performance compared with other state-of-the-art methods demonstrates the effectiveness of our proposed method.
Funder
National Natural Science Foundation of China Fundamental Research Funds for the Central Universities
Subject
Electrical and Electronic Engineering,Biochemistry,Instrumentation,Atomic and Molecular Physics, and Optics,Analytical Chemistry
Reference68 articles.
1. Long, J., Shelhamer, E., and Darrell, T. (2015, January 7–12). Fully convolutional networks for semantic segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA. 2. Ronneberger, O., Fischer, P., and Brox, T. (2015, January 5–9). U-net: Convolutional networks for biomedical image segmentation. Proceedings of the Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany. 3. Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014, January 23–28). Rich feature hierarchies for accurate object detection and semantic segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA. 4. Catastrophic interference in connectionist networks: The sequential learning problem;McCloskey;Psychology of Learning and Motivation,1989 5. Goodfellow, I.J., Mirza, M., Xiao, D., Courville, A., and Bengio, Y. (2013). An empirical investigation of catastrophic forgetting in gradient-based neural networks. arXiv.
|
|