MatrixCity: A Large-scale City Dataset for City-scale Neural Rendering and Beyond

Bo Dai, Yixuan Li, Dahua Lin, Lihan Jiang, Zhenzhi Wang, Linning Xu, Yuanbo Xiangli

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. If it links a repository, it is listed below.

Abstract

Figure 1: An illustration of MatrixCity dataset's small city. a) We collected high-quality large-scale city scene images for a city-scale neural rendering benchmark from Unreal Engine 5, capturing various city environments from multiple viewing angles. b) Our flexible environment control can collect data under dynamic environmental factors such as varying lighting and weather conditions. c) Our developed platform also facilitates the extraction of additional properties, including depth maps and normal maps. These features open a wide range of opportunities for future research in city-scale neural rendering.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2309.16553")
get_code_for_paper("2309.16553")
have("2309.16553")

Connect an agent — have() is free.