You don't actually use sigmod for the spherical harmonic coefficients, right?

dcharatan / pixelsplat

[CVPR 2024 Oral, Best Paper Runner-Up] Code for "pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction" by David Charatan, Sizhe Lester Li, Andrea Tagliasacchi, and Vincent Sitzmann

http://davidcharatan.com/pixelsplat/

MIT License

830 stars 56 forks source link

You don't actually use sigmod for the spherical harmonic coefficients, right? #53

Closed caiyongqi closed 6 months ago

caiyongqi commented 6 months ago

https://github.com/dcharatan/pixelsplat/blob/f4ae9c1396ce62097b407e819321f5819d7788a9/src/model/encoder/common/gaussian_adapter.py#L74

dcharatan commented 6 months ago

That's correct—that comment was left over from before we added support for spherical harmonics. Since we're using spherical harmonics now, the values shouldn't be mapped to the range [0, 1], and so there's no sigmoid anymore. Thanks for pointing this out!