引导式测验 · 中等 等级
Fréchet 起始距离测验
查看 FID 的特征空间高斯比较及其样本大小敏感性，以及评估生成图像保真度和多样性的替代方案。
问题 1 的 8
FID 分数较低表明什么？
预览每个问题8 问题
当您参加测验时，就会出现答案和解释。
- FID 分数较低表明什么？
- 生成的图像的统计数据更接近真实图像的统计数据
- 生成的图像不太真实
- 模型训练次数较少
- 图像具有更高的分辨率
- 标准 FID 使用哪个预训练网络来提取图像特征？
- ResNet-50
- 盗梦空间-v3
- VGG-16
- 伯特
- FID在比较每组图像特征之前如何总结它们？
- 作为单个平均像素值
- 作为文件名的排序列表
- 作为具有均值向量和协方差矩阵的多元高斯分布
- 仅作为颜色通道的直方图
- FID 解决了 Inception Score 的哪些主要缺点？
- Inception Score 计算速度太慢
- Inception Score 仅适用于文本
- Inception Score 不需要神经网络
- Inception Score 从未将生成的图像与真实数据进行比较
- When a generator produces less varied Inception features than the real images, how can that affect FID?
- A covariance mismatch can contribute to a higher FID
- FID must become zero
- FID ignores feature diversity
- FID automatically identifies the exact cause of the collapse
- In the Diffusion Models Beat GANs comparison, how was FID evaluated for ImageNet?
- With 500 generated samples
- With 50,000 generated samples
- With 5,000 generated samples
- With 5 million generated samples
- Why can a small evaluation set make an FID estimate unreliable?
- FID requires every image to have a human label
- The Inception network stops extracting features on small sets
- Finite-sample bias can distort the estimated distance between feature distributions
- Small sets force all covariance values to zero
- What does Kernel Inception Distance (KID) offer as an alternative in the guide?
- A method for training the Inception-v3 network
- A way to generate image captions
- A measure of video frame rate
- A metric used for evaluation when sample counts are smaller