You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently we support UV-coordinate quantization only with fixed factor to reduce storage to e.g. 16 bits per channel. However, this limits both precision and UV-range. By calculating UV-bounds (for each UV-channel separately) and having those bounds available for quantization (in vfmt.xml similar to mesh_bvol_pos and mesh_bvol_rad) would optimize the precision for any UV-range.
The text was updated successfully, but these errors were encountered:
Currently we support UV-coordinate quantization only with fixed factor to reduce storage to e.g. 16 bits per channel. However, this limits both precision and UV-range. By calculating UV-bounds (for each UV-channel separately) and having those bounds available for quantization (in vfmt.xml similar to
mesh_bvol_pos
andmesh_bvol_rad
) would optimize the precision for any UV-range.The text was updated successfully, but these errors were encountered: