知识卡片
分类条件分布逐步组成图像
8位灰度单像素可用256类分布完整表示,给定已见前缀后再用条件分类器预测下一个像素。RGB也可按红、绿、蓝顺序预测,让后通道读取先通道。输出应为条件分布,而非只返回一个最可能灰度。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 生成模型
原书出处 · 1 处
- · 查看原书依据
These models can be easily understood by first considering the problem of synthesizing one pixel, then two, and so on. The first observation to make is that it's pretty easy to synthesize a single grayscale pixel. Such a pixel can take on 256 possible values (for a standard 8-bit grayscale image). So it suffices to use a categorical distribution to represent the probability that the
参考来源
- 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第32章《生成模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c032.xhtml)
- 结论依据:摘录中关于分类条件分布逐步组成图像的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。
- 原始内容:These models can be easily understood by first considering the problem of synthesizing one pixel, then two, and so on. The first observation to make is that it's pretty easy to synthesize a single grayscale pixel. Such a pixel can take on 256 possible values (for a standard 8-bit grayscale image). So it suffices to use a categorical distribution to represent the probability that the