DualFusionNet: Comparison of Computer Vision Model Performance from Cross-Validation

A comparison table of performance metrics for the DualFusionNet architecture against ResNet50 and Swin Transformer V2 models. The dataset contains results averaged from five runs of cross-validation. It was authored by Christopher Mai and last updated on April 29, 2026.

Use Cases

• Benchmarking the DualFusionNet architecture against established models like ResNet50 and Swin Transformer V2 based on the described comparison.
• Analyzing the impact of the Squeeze-and-Excitation (SE) block by comparing DualFusionNet with and without it.
• Evaluating the performance of the Adaptive Feature Pyramid Network (AFPN) within a fusion architecture.
• Assessing model stability and generalization using results averaged from five cross-validation runs.

Strengths

• Results are averaged from five cross-validation runs, which may provide a more stable performance estimate.
• Explicitly compares multiple modern architectures (ResNet50, Swin Transformer V2, DualFusionNet variants).
• Released under a permissive CC-BY-4.0 license for clear reuse.

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

figshare

Collection Method

Likely contains experimental results from model training and evaluation.

Time Range

null

Freshness

Last updated 2026-04-29 17:50:50; freshness should be verified.

Geography

null