中心在自监督学习、元学习、神经网络结构优化、知识蒸馏等深度学习方向不断深耕。在小样本学习方面,从基于优化的元学习视角出发,提出了梯度优化改进算法,为基于模型无关的元学习提供了新的改进路线。近年来随着深度学习技术发展迅速,中心致力于融合自监督学习和知识蒸馏方法,增强模型对特征的表达能力,以期达到更好的性能和精度,更好地完成各类下游任务等,相关工作发表于CVPR等顶级国际会议和期刊上。
代表性论文成果:
[1] Chao Li, Yixiao Yang, Kun He, Stephen Lin, John E. Hopcroft. Single Image Reflection Removal Through Cascaded Refinement. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020.
[2] Shuoxi Zhang, Hanpeng Liu, Stephen Lin, Kun He. You Only Need Less Attention Each Stage on Vision Transformers, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2024.
[3] Chen Q, Li C, Ning J, et al. GMConv: Modulating Effective Receptive Fields for Convolutional Kernels[J]. IEEE Transactions on Neural Networks and Learning Systems, 2024.