Closed atharvadeore999 closed 1 month ago
No, the pretrained models are only for parsing texts. To parse images, you need to train the model yourself. There are image annotations in this dataset while I didn't perform related experiments.
Can you please help me by suggesting some good models that will help me generate scene graphs from images? Your help will be greatly appreciated
Sorry, I did some literature survey two years ago. By that time, none of the image scene graph parsers can work. They just generated a lot of noise. And this is also why I included the image annotation but I didn’t proceed. Because I feel the text parser at least can have useful downstream applications while I couldn’t know what the application would be with a poorly performed image scene graph parser.
However, that was before the era of LLMs, not sure whether the newest multimodal LLM can solve the problem. Or whether the image scene graph parsing problem is still worth solving?
atharvadeore999 @.***>于2024年6月27日 周四上午12:44写道:
Can you please help me by suggesting some good models that will help me generate scene graphs from images? Your help will be greatly appreciated
— Reply to this email directly, view it on GitHub https://github.com/zhuang-li/FactualSceneGraph/issues/6#issuecomment-2191891889, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACTSNJ36LGXXN2YDIFIW5T3ZJLHTPAVCNFSM6AAAAABJ4BO6SOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCOJRHA4TCOBYHE . You are receiving this because you commented.Message ID: @.***>
Thank you so much for letting me know
On Thu, Jun 27, 2024 at 7:40 PM Zhuang Li @.***> wrote:
Closed #6 https://github.com/zhuang-li/FactualSceneGraph/issues/6 as completed.
— Reply to this email directly, view it on GitHub https://github.com/zhuang-li/FactualSceneGraph/issues/6#event-13318183894, or unsubscribe https://github.com/notifications/unsubscribe-auth/ATV6GZ3PDDN47JYOZSUSM2LZJQMMDAVCNFSM6AAAAABJ4BO6SOVHI2DSMVQWIX3LMV45UABCJFZXG5LFIV3GK3TUJZXXI2LGNFRWC5DJN5XDWMJTGMYTQMJYGM4DSNA . You are receiving this because you authored the thread.Message ID: @.***>
Can we use the pretrained model on images to generate scene graph .If yes please let me know the steps required to perform it.