search

alekssamos / cloudvision

Vision Bot NVDA addon
https://visionbot.ru/addon/
MIT License
9 stars 9 forks source link

CloudVision, any chance of extending it to recognize the current screen or window? #15

Open amirsol81 opened 6 months ago

amirsol81 commented 6 months ago

@alekssamos First and foremost, thanks for fixing the Be My AI issue in V3.2.0.1. Now that Be My AI is available, would it be possible to extend CloudVision's functionality in a way that it can become capable of recognizing the currently focused window or screen? This would help a lot with, say, opened images in Telegram, inaccessible apps which display text which cannot be detected via NVDA's cursor/object navigation key strokes, etc. Maybe new hot keys can be added to cover screen/window detection/recognition. The current approach is either object-based or file-based, but my suggestion can expand the usefulness of the add-on. Thanks.

alekssamos commented 6 months ago

Yes, okay, maybe I will. But I can not promise. Usually I just go up to the top level through object navigation and that’s it. I always do this automatically and without even thinking.

amirsol81 commented 6 months ago

@alekssamos If doable and if it doesn't bother you, it would be both hugely beneficial and appreciated!

amirsol81 commented 4 months ago

@alekssamos Greetings. Any chance of working on this to cover window and screen recognition?