MiDaS-DPT vs. Pothole-Specific Depth Estimation Models: Performance Metrics

5.5 KB of performance metrics comparing the MiDaS-DPT model against state-of-the-art pothole-specific methods for depth estimation. The dataset, authored by Waqar Rauf Butt and last updated on 2026-05-08, includes evaluation metrics such as MAE, RMSE, and accuracy. It is shared on figshare under a CC-BY-4.0 license.

Use Cases

• Benchmarking depth estimation models for pothole detection based on the described MAE, RMSE, and accuracy metrics.
• Comparing general-purpose depth models (like MiDaS-DPT) against task-specific methods based on the performance evaluation.
• Informing model selection for autonomous vehicle or road maintenance systems based on the comparative analysis of depth estimation accuracy.

Strengths

• Dataset is small (5.5 KB), facilitating quick download and inspection.
• Includes a direct comparison of a specific model (MiDaS-DPT) against pothole-specific methods, providing a focused benchmark.
• Uses standard evaluation metrics (MAE, RMSE, accuracy) for depth estimation.

Limitations

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

Provenance

Source

figshare

Collection Method

Likely contains results from experimental evaluation or benchmarking.

Freshness

Last updated 2026-05-08 17:54:12; freshness should be verified.