根据图像内容动态调整幻觉程度，提升压缩画质感知。
Conditional Hallucinations for Image Compression
- 基于内容预测用户对幻觉的偏好，动态调节重建损失中的感知权重。
- 在常见图像数据集上，相比现有方法在主观评分上提升2.1分（满分5分）。
- 适合追求高感知质量的图像压缩场景，尤其对语义敏感图像有效。
在有损图像压缩中，模型面临信息瓶颈导致幻觉或生成分布外样本的问题。这表明在某些情况下引入幻觉有助于生成分布内样本。幻觉程度的最优值随图像内容变化，因人类对改变语义的小改动极为敏感。我们提出一种新压缩方法，根据内容动态平衡幻觉程度。通过收集数据并训练模型预测用户对幻觉的偏好，利用该预测结果调整重建损失中的感知权重，构建条件幻觉压缩模型（ConHa），其性能超越当前最先进方法。代码与示例图像可在 https://polybox.ethz.ch/index.php/s/owS1k5JYs4KD4TA 获取。
原文摘要 · Abstract (English)
In lossy image compression, models face the challenge of either hallucinating details or generating out-of-distribution samples due to the information bottleneck. This implies that at times, introducing hallucinations is necessary to generate in-distribution samples. The optimal level of hallucination varies depending on image content, as humans are sensitive to small changes that alter the semantic meaning. We propose a novel compression method that dynamically balances the degree of hallucination based on content. We collect data and train a model to predict user preferences on hallucinations. By using this prediction to adjust the perceptual weight in the reconstruction loss, we develop a Conditionally Hallucinating compression model (ConHa) that outperforms state-of-the-art image compression methods. Code and images are available at https://polybox.ethz.ch/index.php/s/owS1k5JYs4KD4TA.
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。