gltfpack: Preserve custom ID attributes #568
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This change keeps attributes with _ID_ or _ID in the name in tact when processing the scene. We assume that these attributes are scalar and encode them as floating-point (to avoid possible range issues with uint16); the filter doesn't seem to be necessary here although we could use an exponential filter for slightly better compression in the future.
This results in ~2x smaller attribute streams so it is very worthwhile; as with other compression filters, we only apply this when -cc is specified. The encoding assumes that the values are in 0..2^23-1 range; since input values may be encoded as floats which permits a slightly wider range (integers up to 2^24 are encoded exactly), we guard against this to make sure no data loss occurs.
It should be fine to use custom_name whenever it's specified; it is only assigned for _custom attributes right now anyhow.
The previous search would not detect _BATCHID (Cesium 3D Tiles) as valid, and would not detect some odd strings like _IDEAL_ID. For now let's simply recognize the attributes that are known to be used in the ecosystem as IDs; we can expand this list or replace it with a different identification strategy in the future.
Closed
|
Btw. we've noticed that even though |
|
Yes, gltfpack must implement extensions for them to be preserved, as is common with other glTF processing tools. In case of EXT_mesh_features specifically, this extensions doesn't really exist yet in that it has been proposed (there's an open PR on glTF repository) but has not been merged or ratified, so I haven't looked into what it would take to support it. |
|
Got it. Thanks @zeux! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This change keeps attributes
_ID,_BATCHIDand_FEATURE_ID_*in tact when processing the scene. We assume that these attributes are scalar and encode them as floating-point (to avoid possible range issues with uint16).This change is similar to #530 (without the extraneous names) but it crucially limits the supported attributes by name otherwise we can not assume that the attribute is scalar or integer, whereas with this filtering the assumption is reasonably safe.
For increased compression efficiency, when
-ccis applied we use exponential encoding unless the IDs exceed the maximum permissible range. On some test files that use feature and vertex ids, this reduces the compressed size of ID attribute streams by ~2x.Relates to #165
Fixes #529
Closes #530