GraspClutter6D: Real-World Robotic Grasping Data for Cluttered Scenes

A curated subset of 99 scenes from the GraspClutter6D dataset, hosted as a FiftyOne dataset with 111 samples. It is a large-scale real-world dataset for robust perception and grasping in cluttered scenes, created by Voxel51 and last updated in May 2026.

Use Cases

• Training object detection models for cluttered scenes based on the real-world imagery.
• Developing and benchmarking robotic grasping algorithms based on the dataset's focus on robust grasping.
• Visualizing and exploring perception challenges in complex environments using the FiftyOne platform.

Strengths

• Contains 99 curated scenes from a large-scale dataset.
• Optimized for visualization and exploration within the FiftyOne platform.
• Focuses on real-world cluttered scenes, a challenging domain for robotics.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Voxel51

Freshness

Last updated 2026-05-26 18:36:12; freshness should be verified.