Hi, nice work.
I was wondering whether the model can perform inference on large images. In real-world applications, images are often much larger, such as 1024 × 1920. Since the model in this repository seems to be limited to 512 × 512 inputs, is overlapping tile-based inference the recommended approach for handling large images?
Hi, nice work.
I was wondering whether the model can perform inference on large images. In real-world applications, images are often much larger, such as 1024 × 1920. Since the model in this repository seems to be limited to 512 × 512 inputs, is overlapping tile-based inference the recommended approach for handling large images?