🔍 𝗖𝗵𝗮𝗿𝗴𝗲 is based on an animated movie providing a variety of motions and high-quality textures

🔍 𝗖𝗵𝗮𝗿𝗴𝗲 introduces 3 benchmarking scenarios for common setups: Dense, Sparse, Monocular

🔍 𝗖𝗵𝗮𝗿𝗴𝗲 offers high-quality RGB images accompanied by depth, segmentation, optical flow, dynamic masks, normals

📣 𝗖𝗵𝗮𝗿𝗴𝗲: A Comprehensive Benchmark and Dataset for Dynamic Novel View Synthesis

TL;DR: A new dataset and benchmark for Dense, Sparse and Monocular reconstruction.

🔗 charge-benchmark.github.io

M. Nazarczuk, T. Tanay @thomas-tanay.bsky.social, Z. Zhang, E. Pérez-Pellitero @eduperezp.bsky.social

👇