We propose 4Real, the first photorealistic text-to-4D scene generation pipeline.

We present CoGS, a method for Controllable Gaussian Splatting, that enables the direct manipulation of scene elements, offering real-time control of dynamic scenes without the prerequisite of pre-computing control signals.

We propose propose the Dynamic Light Field Network (DyLiN) method that can handle non-rigid deformations, including topological changes, which outperformed state-of-the art methods in terms of visual fidelity and compute complexity.

We propose a general adversarial learning framework for solving Unsupervised 2D to Explicit 3D Style Transfer.

We propose a fully-automatic controllable neural representation for face self-portraits.

We propose a parallel network framework and introduce an attention mechanism to improve subspace-based zero-shot self-supervised learning and enable higher acceleration factors.

We develop a scan-specific model that estimates and corrects k-space errors made when reconstructing accelerated MRI data.

We accelerate RAKI by more than 200 times by directly learning a coil-combined target.

We present a multi-task deep learning approach that allows simultaneous tumor segmentation and response prediction of pathologic complete response after neoadjuvant chemoradiotherapy.

We use mixed kernels to improve the performance of existing medical image segmentation networks.

We propose a novel deep neural network (S-net) to construct a CT signature for predicting disease-free survival (DFS) and overall survival.

We develop a deep learning system for predicting lymph node metastasis in multiple nodal stations based on preoperative CT images in patients with gastric cancer.

We develop an image-based person detection algorithm for wearable computing using commodity smartphones.

We propose a robust method for automatic left atrial appendage segmentation on computed tomographic angiography data using fully convolutional neural networks with 3D conditional random fields.

we present a new approach for the detection of substances in the left atrial appendage by spatiotemporal motion analysis and make a detailed judgment and analysis of spatial distribution and classification of most objects in the left atrial appendage.

We propose a robust method for automatic left atrial appendage segmentation on computed tomographic angiography data using fully convolutional neural networks with 3D conditional random fields.