Abstract: Semantic image synthesis aims to generate high-quality images given semantic conditions, i.e., segmentation masks and style reference images. Existing methods widely adopt generative ...

Abstract: As one of the core tasks in vision recognition, image classification is widely used in various scenarios. Most existing mainstream image classification models use the Convolutional Neural ...

A lightweight Python deep learning framework for precision agriculture. It leverages a CNN autoencoder to detect mixed pixels in thermal images, enabling early crop disease detection with robust ...

I deployed a super-resolution model on the Android side using TensorFlow, with a runtime of 90ms (540p ->1080p) when calling QNN. However, the preprocessing and post-processing of the images were slow ...