DriveScape: Towards High-Resolution Controllable Multi-View Driving Video Generation

no code implementations • 9 Sep 2024 • Wei Wu, Xi Guo, Weixuan Tang, Tingxuan Huang, Chiyu Wang, Dongyue Chen, Chenjing Ding

However, existing approaches often struggle with multi-view video generation due to the challenges of integrating 3D information while maintaining spatial-temporal consistency and effectively learning from a unified model.

Autonomous Driving Video Generation

Mask-adaptive Gated Convolution and Bi-directional Progressive Fusion Network for Depth Completion

no code implementations • 15 Jan 2024 • Tingxuan Huang, Jiacheng Miao, Shizhuo Deng, Tong, Dongyue Chen

Depth completion is a critical task for handling depth images with missing pixels, which can negatively impact further applications.

Decoder Depth Completion

AGG-Net: Attention Guided Gated-convolutional Network for Depth Image Completion

1 code implementation • ICCV 2023 • Dongyue Chen, Tingxuan Huang, Zhimin Song, Shizhuo Deng, Tong Jia

In the encoding stage, an Attention Guided Gated-Convolution (AG-GConv) module is proposed to realize the fusion of depth and color features at different scales, which can effectively reduce the negative impacts of invalid depth data on the reconstruction.