CAN-Age: Experimental Results for Color-Attention Based Wildlife Age Estimation

An Excel file containing experimental results and evaluation metrics for the paper "CAN-Age: Color-Attention Based Deep Learning Architecture for Noninvasive Wildlife Age Estimation". The dataset, authored by Yincalos Peng, was last updated on April 27, 2026. It records predictive performance across four backbone architectures and three ablation settings.

Use Cases

• Comparing the performance of four deep learning architectures (Standard CNN, VGG, ResNet, DenseNet) for age prediction tasks.
• Evaluating the effectiveness of a proposed Color-Attention mechanism against baseline and color-ablated models.
• Analyzing model performance metrics (True Age vs. Predicted Age) across different experimental configurations.

Strengths

• Provides a structured comparison across 12 distinct experimental sheets.
• Includes results for three specific ablation settings (Baseline, NoColor, CA) per architecture.
• Released under a permissive CC-BY-4.0 license.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small (57.2 KB), indicating limited scope.

Provenance

Source

figshare, authored by Yincalos Peng.

Collection Method

Likely contains experimental results from the research paper's evaluation phase.

Time Range

null

Freshness

Last updated 2026-04-27 19:26:01; freshness should be verified.

Geography

null