Producing and Leveraging Online Map Uncertainty in Trajectory Prediction

Abstract

We present SPAD, a novel approach for creating consistent multi-view images from text prompts or single images. To enable multi-view generation, we repurpose a pretrained 2D diffusion model by extending its self-attention layers with cross-view interactions, and fine-tune it on a high quality subset of Objaverse. We find that a naive extension of the self-attention proposed in prior work (e.g. MVDream) leads to content copying between views. Therefore, we explicitly constrain the cross-view attention based on epipolar geometry. To further enhance 3D consistency, we utilize Plucker coordinates derived from camera rays and inject them as positional encoding. This enables SPAD to reason over spatial proximity in 3D well. In contrast to recent works that can only generate views at fixed azimuth and elevation, SPAD offers full camera control and achieves state-of-the-art results in novel view synthesis on unseen objects from the Objaverse and Google Scanned Objects datasets. Finally, we demonstrate that text-to-3D generation using SPAD prevents the multi-face Janus issue.

Publication
Conference on Computer Vision and Pattern Recognition (CVPR)

Toronto Intelligent Systems Lab Co-authors

Yash Kant
Yash Kant
PhD Student

I enjoy talking to people and building (hopefully useful) things together. :)

Ziyi Wu
Ziyi Wu
PhD Student

Hi! I am a PhD student working on computer vision. My research interests include representation learning, 3D vision, and event-based vision.

Igor Gilitschenski
Igor Gilitschenski
Assistant Professor