pytorch/vision

torchvision.transforms.functional.adjust_hue does not respect alpha channel

Open

#3,219 建立於 2021年1月4日

在 GitHub 查看
 (5 留言) (0 反應) (0 負責人)Python (15,050 star) (6,858 fork)batch import
enhancementhelp wantedmodule: documentationmodule: transforms

描述

🐛 Bug

when working with RGBA images, the adjust_hue function does not seem to respect the alpha channel and imposes a max_value of 255 after the transform. Taking a quick look at the source code, it seems that the image is immediately converted to HSV without retaining the alpha channel. It should be a quick fix to retain the alpha channel and include it when merging back into RGBA.

To Reproduce

Steps to reproduce the behavior:

img = Image.open('xyz.png')
img_ = adjust_hue(img, 0.1)
print(np.array(img.split()[-1]).mean())
print(np.array(img_.split()[-1]).mean())

cc @vfdev-5

貢獻者指南

技術棧
pythonpytorch
領域
backendapi
議題類型
bug
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
2
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
under 1 hour
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
stale
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
clear
前置要求
PythonPyTorchImage processing
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
80
研究方向
Look at the adjust hue function in torchvision/transforms/functional.py. The issue is that it converts RGBA to HSV without saving the alpha channel, then reconstructs without alpha. A fix is to split the alpha channel before conversion and merge it back after adjusting hue. Also check if adjust saturation or adjust contrast have similar issues.