Comparison of Different Loss Functions for Smoking Detection in Construction Sites

A 5.5 KB Excel file by Ruishi Liang, last updated on 2026-05-06, comparing novel loss functions for object detection models. The dataset likely contains performance metrics from experiments evaluating the AOW-YOLO model against other lightweight models like PP-LCNet and YOLO11n. The research focuses on detecting smoking behavior in complex construction site environments.

Use Cases

• Benchmarking new loss functions like AOWIoU against established metrics based on the described experimental results.
• Evaluating model inference speed improvements for deployment on edge devices based on the reported 31.6% speed gain.
• Training or fine-tuning object detection models for safety compliance in cluttered environments based on the described smoking detection application.

Strengths

• Explicitly reports a 31.6% faster inference speed than YOLO11n in the model evaluation.
• Provides a direct comparison of mAP50 and mAP50:95 metrics for model performance.
• Released under a permissive CC-BY-4.0 license for open use and modification.

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.
• The 5.5 KB file size suggests a very limited scope, likely containing summary results rather than raw training data.

Provenance

Source

figshare

Collection Method

Likely contains experimental results from model training and evaluation, as described in the research summary.

Time Range

null

Freshness

Last updated 2026-05-06 17:45:43; freshness should be verified.

Geography

null