用数学序列优化图像生成顺序，让画图更清晰更高效。
Halton Scheduler For Masked Generative Image Transformer
- 用哈尔顿序列替代原采样策略，按空间均匀分布选择像素点。
- 在ImageNet和COCO上FID降低，生成图像更细致多样。
- 无需重训练或改模型，可直接替换原方案，适合快速部署。
掩码生成图像变换器（MaskGIT）作为高效可扩展的图像生成框架，能以低推理成本生成高质量图像。然而其关键组件——掩码解码调度器尚未得到足够重视。本文分析了MaskGIT中的采样目标，基于标记间的互信息揭示其不足。提出基于准随机、低差异哈尔顿序列的新采样策略，使被选位置在空间上更均匀分布，逐步覆盖图像。该方法可减少不可恢复的采样误差，简化超参数调优，并提升图像质量。该调度器无需重新训练或注入噪声，可作为即插即用的替代方案。在ImageNet类条件生成和COCO文本到图像生成任务上的评估表明，相比原有置信度调度器，哈尔顿调度器在定量指标（FID降低）和定性效果（图像更丰富、细节更清晰）上均表现更优。
原文摘要 · Abstract (English)
Masked Generative Image Transformers (MaskGIT) have emerged as a scalable and efficient image generation framework, able to deliver high-quality visuals with low inference costs. However, MaskGIT's token unmasking scheduler, an essential component of the framework, has not received the attention it deserves. We analyze the sampling objective in MaskGIT, based on the mutual information between tokens, and elucidate its shortcomings. We then propose a new sampling strategy based on our Halton scheduler instead of the original Confidence scheduler. More precisely, our method selects the token's position according to a quasi-random, low-discrepancy Halton sequence. Intuitively, that method spreads the tokens spatially, progressively covering the image uniformly at each step. Our analysis shows that it allows reducing non-recoverable sampling errors, leading to simpler hyper-parameters tuning and better quality images. Our scheduler does not require retraining or noise injection and may serve as a simple drop-in replacement for the original sampling strategy. Evaluation of both class-to-image synthesis on ImageNet and text-to-image generation on the COCO dataset demonstrates that the Halton scheduler outperforms the Confidence scheduler quantitatively by reducing the FID and qualitatively by generating more diverse and more detailed images. Our code is at https://github.com/valeoai/Halton-MaskGIT.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。