DCVerse: Unified Benchmarks for Depth Completion

DCVerse is a unified benchmark for depth completion methods. It standardizes experimental settings to facilitate fair and reproducible evaluation. The benchmark was created by Liangyingping and last updated on June 9, 2026.

Use Cases

• Benchmark depth completion algorithm performance based on unified input image resolution.
• Compare method accuracy based on consistent sparse point sampling density.
• Evaluate model generalization across different datasets using standardized experimental settings.

Strengths

• Addresses inconsistencies in prior evaluations, including unified input image resolution.
• Provides standardized sparse point sampling density for consistent comparisons.
• Last updated on 2026-06-09, suggesting recent maintenance.

Limitations

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

Provenance

Source

huggingface

Freshness

Last updated 2026-06-09 09:04:10.