MVPbev: Multi-view Perspective Image Generation from BEV with Test-time Controllability and Generalizability

TL;DR

MVPbev is a novel method for generating multi-view, photorealistic images from text prompts using BEV semantics, with test-time controllability and improved generalization to unseen views.

Contribution

The paper introduces MVPbev, a two-stage model that projects BEV semantics to perspective views and enforces cross-view consistency, enabling controllable and generalizable multi-view image generation from text.

Findings

Outperforms state-of-the-art on NuScenes dataset

Generates high-resolution photorealistic images from text

Demonstrates strong generalization to unseen viewpoints

Abstract

This work aims to address the multi-view perspective RGB generation from text prompts given Bird-Eye-View(BEV) semantics. Unlike prior methods that neglect layout consistency, lack the ability to handle detailed text prompts, or are incapable of generalizing to unseen view points, MVPbev simultaneously generates cross-view consistent images of different perspective views with a two-stage design, allowing object-level control and novel view generation at test-time. Specifically, MVPbev firstly projects given BEV semantics to perspective view with camera parameters, empowering the model to generalize to unseen view points. Then we introduce a multi-view attention module where special initialization and de-noising processes are introduced to explicitly enforce local consistency among overlapping views w.r.t. cross-view homography. Last but not least, MVPbev further allows test-time instance-level controllability by refining a pre-trained text-to-image diffusion model. Our extensive experiments on NuScenes demonstrate that our method is capable of generating high-resolution photorealistic images from text descriptions with thousands of training samples, surpassing the state-of-the-art methods under various evaluation metrics. We further demonstrate the advances of our method in terms of generalizability and controllability with the help of novel evaluation metrics and comprehensive human analysis. Our code, data, and model can be found in \url{https://github.com/kkaiwwana/MVPbev}.