IEEE

Learning Many-to-Many Mapping for Unpaired Real-World Image Super-Resolution and Downscaling

Abstract: Learning based single image super-resolution (SISR) for real-world images has been an active research topic yet a challenging task, due to the lack of paired low-resolution (LR) and ...
IEEE

Automatic Liver Tumor Segmentation on Dynamic Contrast Enhanced MRI Using 4D Information: Deep Learning Model Based on 3D Convolution and Convolutional LSTM

Abstract: Objective: Accurate segmentation of liver tumors, which could help physicians make appropriate treatment decisions and assess the effectiveness of surgical treatment, is crucial for the ...
GitHub

TesserAct: Learning 4D Embodied World Models

We propose TesserAct, the first open-source and generalized 4D World Model for robotics, which takes input images and text instructions to generate RGB, depth, and normal videos, reconstructing a 4D ...