Add Pheno4D point cloud segmentation dataset - #62
Merged
Merged
Conversation
Follows the EmbeddingPlot3D pattern (react-plotly.js, scatter3d), adapted for raw point coordinates colored by mask value instead of embedding-space coordinates colored by class. Wired into the Sample image section rather than the benchmark/embeddings view, since that pipeline is built for UMAP embeddings and doesn't apply to point cloud data. Tested locally against the live Pheno4D dataset card.
Supports datasets covering multiple species (e.g. Pheno4D's maize and tomato) by restructuring the sample file into named per-crop entries instead of one flat points array. A selector toggle (mirroring the existing 2D/3D embedding toggle pattern) appears when more than one crop is present. Camera resets on crop switch. Sample data updated with real maize (M01_0313_a) and tomato (T01_0305_a) scans, both dual and single label schemes preserved.
js2552
approved these changes
Sep 12, 2026
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds Pheno4D point cloud segmentation dataset
First point cloud dataset added to AgML. Live on Hugging Face: https://huggingface.co/datasets/Project-AgML/pheno4d_point_cloud_segmentation (126 labeled scans, maize + tomato).
What's in this PR
static/data/hf_datasets.jsonNote on loader compatibility
This dataset's
mask/mask_2columns areArray2D(variable-length numeric arrays), not theImage()typeHuggingFaceDataLoadercurrently expects for segmentation. So while this entry renders correctly on the site (using the closest valid enum values:machine_learning_task: semantic_segmentation,annotation_format: segmentationMask), the dataset likely won't load correctly through the existing loader without changes there. So it would need a new loader path.