no code implementations • 13 Nov 2022 • Yan Luo, Yangcheng Gao, Zhao Zhang, Haijun Zhang, Mingliang Xu, Meng Wang
We find it is because: 1) a normal generator is hard to obtain high diversity of synthetic data, since it lacks long-range information to allocate attention to global features; 2) the synthetic images aim to simulate the statistics of real data, which leads to weak intra-class heterogeneity and limited feature richness.
no code implementations • 30 Apr 2022 • Yangcheng Gao, Zhao Zhang, Richang Hong, Haijun Zhang, Jicong Fan, Shuicheng Yan
To obtain high inter-class separability of semantic features, we cluster and align the feature distribution statistics to imitate the distribution of real data, so that the performance degradation is alleviated.