Nguyen Xuan Hung

Hi, react native is a cross-platform, it will be missing if we only support IOs... I havn't found any solution to convert .heic for bolt android and ios yet

@cliffordkleinsr @aliagha1997 @weizmann @cliffordkleinsr. We can modify the code to batch inference, here is my simple code with batch_size = 2: ```python from PIL import Image import time import cv2...

`predictor.set_image(image)` is the step generates image embedding > To use the ONNX model, the image must first be pre-processed using the SAM image encoder \- this is the main thread...

> @hungtooc Thanks for your comment, For me, other processes including pre-processing and transofrmation does not take that much long, and most of the time is taken at the line...

@KAWAKO-in-GAYHUB I not sure, maybe he could help you https://github.com/facebookresearch/segment-anything/issues/107#issuecomment-1500909850

> input_label = np.array([1, 1]) you have 3 points, you much have 3 labels too ```python input_label = np.array([1, 1, 1]) ```

> BTW, is it possible to specify multiple prompts for different labels for multiple classes? For example, > > input_point = np.array([[500, 1375], [1500, 1375], [1125, 1625], [482, 923], [1832,...

Yes, checkout https://github.com/UX-Decoder/Segment-Everything-Everywhere-All-At-Once. they combine models together to solve problem

yes, we need code to create custom mask

Update, I guess they preprocess mask by logits: ```python def logit(x): a = torch.tensor(x) return torch.special.logit(a, eps=1e-6).numpy() def inv_logit(x): return 0.5*(1. + np.sign(x)*(2./(1. + np.exp(-np.abs(x))) - 1.)) #load mask mask...