Replies: 1 comment 1 reply
-
|
作者还有兴趣吗？我正在做动态检测圣遗物框位置，目前是为了让playcover的兼容性更好一点。 另外个人感觉这种识别的目标其实很有限，未必需要深度学习法案，Bag-of-words应该就行。 |
Beta Was this translation helpful? Give feedback.
1 reply
-
|
很欢迎动态框位置检测，bow可以试试，要是效果更好也能考虑 |
Beta Was this translation helpful? Give feedback.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
👋 Yas未来计划支持扫描背包、角色等数据
但是作者并不熟CV，特来探讨可行的深度学习方案，或者传统CV方案
目前想到的背包识别方案：
如何提高CRNN准确率，或者其他的识别方案
目前是MobileNetV3_Small+LSTM
Beta Was this translation helpful? Give feedback.
All reactions