search

phuang17 / DeepMVS

DeepMVS: Learning Multi-View Stereopsis
https://phuang17.github.io/DeepMVS/index.html
BSD 2-Clause "Simplified" License
331 stars 85 forks source link

The create of MVS-Synth dataset #16

Closed rann1018 closed 4 years ago

rann1018 commented 4 years ago

Hi,

Thanks for your work and open-source MVS-Synth dataset. I read your paper and am aware that you created this dataset from GTA5. However, how can you get the ground truth disparity maps, and the extrinsic and intrinsic camera parameters. I love this game but I think it can only capture images or videos. Hope you can lend me some advice.

Thanks a lot.

phuang17 commented 4 years ago

@rann1018 I used RenderDoc to access the Z-buffer, which can then be converted to depth maps. The extrinsic and intrinsic camera parameters can also be computed by the model-to-world matrix and the model-to-world-to-view matrix withing the RenderDoc buffer results. It will require some knowledge about Computer Graphics. Please let me know if you have any other questions. Thanks!