It can be observed that features with low resolutions (e.g., Figure 2 shows the semantic interpolation results. Figure 1: Visualization of the semantics learned in different resolutions. It can be observed that features with low resolutions (i.e., For clear representation, the tags (i.e., glasses eyes, mouths, and beard) are SariGAN (semantic-specific control) can control a specific semantic while preserving identities.

Summary and Contributions: This paper improves the StyleGAN-based image generation model by disentangling semantics based on a learnable semantics grouping operation, where the styles of the intra-group features are controlled by group-wise adaptive instance normalization and the overall features are re-balanced by inter-group adaptive group normalization. Quantitative and qualitative evaluations show certain improvements over existing methods. Strengths: - The quantitative evaluations and ablation study validates the effectiveness of the proposed improvements. The most critical limitation of this work is its novelty and theoretical soundness. However, similarity between layers of a convolutional kernel may not indicate consistent similarity between corresponding feature channels.

R3 and R4 rate the paper top 50% papers, while R1 votes the paper marginally below the bar. While R1 initially raised several concerns on the paper's novelty side, R1 upgrades the rating of the paper since the rebuttal addresses the concerns. After consolidating the reviews and rebuttal, the AC finds the proposed method interesting. The channel grouping and normalization based on the filter similarity is new for generator design, and the results and analysis presented in the paper support the claim. The AC determines that the paper has merits to be published in the NeurIPS conference and would like to recommend its acceptance.

The recent advances in image generation have been achieved by style-based image generators. Such approaches learn to disentangle latent factors in different image scales and encode latent factors as "style" to control image synthesis. However, existing approaches cannot further disentangle fine-grained semantics from each other, which are often conveyed from feature channels. In this paper, we propose a novel image synthesis approach by learning Semantic-aware relative importance for feature channels in Generative Adversarial Networks (SariGAN). Particularly, we learn to cluster feature channels by semantics and propose an adaptive group-wise Normalization (AdaGN) to independently control the styles of different channel groups.