NeRFLiX: High-Quality Neural View Synthesis by Learning a Degradation-Driven Inter-viewpoint MiXer

Xiaoguang Han, Jiangbo Lu, Kun Zhou, Yi Wang, Wenbo Li, Sse, Nianjuan Jiang, Tao Hu

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.

Repositories linked to this paper

redrock303/NeRFLiX_CPVR2023 — marked official in Papers with Code

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Figure 1. We propose NeRFLiX, a general NeRF-agnostic restorer that is capable of improving neural view synthesis quality. The first example is from Tanks and Temples [25], the second/third examples are from LLFF [36], and the last one is a user scene captured by a mobile phone. RegNeRF-V3 [39] means the model trained with three input views.

For agents

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

get_harvested_code_for_paper("2303.06919")
get_code_for_paper("2303.06919")
have("2303.06919")

Connect an agent — have() is free.