TransitNet Model Parameter Counts and Performance Metrics

Haiyan Zhang published a comparison of model parameter counts and performance metrics for semantic segmentation networks on figshare in May 2026. The dataset likely contains tabular data comparing the proposed TransitNet model against PSPNet and CM-UNet on VOC2012, LoveAD, and OST300 datasets. It includes metrics such as mIoU scores and parameter counts, supporting research into lightweight computer vision models.

Use Cases

• Benchmarking model efficiency based on parameter counts and mIoU scores mentioned in the description
• Evaluating cross-domain adaptation capabilities for traffic and remote sensing scenarios
• Comparing the performance of PSPNet, CM-UNet, and TransitNet architectures

Strengths

• Includes specific performance metrics: TransitNet achieved an mIoU of 86.98% on VOC2012, 61.04% on LoveAD, and 76.11% on OST300
• Dataset is openly licensed under CC-BY-4.0
• Associated with a published research paper and code repository

Limitations

• Dataset size is only 5.5 KB, indicating a very limited scope of data
• Row count and column-level documentation are unknown, which may limit suitability assessment
• Description metadata is limited; actual data quality requires manual inspection after download

Provenance

Source

figshare

Freshness

Last updated 2026-05-22 17:23:51; freshness should be verified