Matting Anything Model (MAM), an efficient and versatile framework for estimating the alpha matte of any instance in an image with flexible and interactive visual or linguistic user prompt guidance.
Hello! Thankyou for sharing your work. I notice that though the performance of MAM is good enough, I can not further perfect my matting result. For example, I want to get an alpha matte of a man with a wine glass on his hand, and MAM can successfully matte the man and the glass separately. But the result may goes bad when I select them together. Can there be an opportunity to add another object to the finished result?
Thank you in advance for your reply!
Well, I find that the merge mode support multiple object and adding object can be easy. But there still can be a function to further modify the results.
Hello! Thankyou for sharing your work. I notice that though the performance of MAM is good enough, I can not further perfect my matting result. For example, I want to get an alpha matte of a man with a wine glass on his hand, and MAM can successfully matte the man and the glass separately. But the result may goes bad when I select them together. Can there be an opportunity to add another object to the finished result? Thank you in advance for your reply!