U-Net · ResNet34 · MARIDA
OverlaysFilters the patch list below. The overlay toggles above are what change the image.
Every overlay, uncertainty map and metric on this page is the real trained model's output, precomputed offline on the MARIDA held-out patches and bundled — no runtime model download.
Keyboard-fast triage of model detections vs MARIDA ground truth
Region-level, scored against MARIDA ground truth across the detections you've reviewed on 0 patches. Updates live.
Corrections accumulate into a retraining set. Low-confidence detections (< op threshold) route here — never auto-accepted.
Drag the cutoff — P/R/F1/IoU recompute live on bundled GT + probs
Computed over — labeled pixels across the — demo patches. Moving the slider also re-thresholds the map overlay in the center. The dot marks the current cutoff on the real PR curve.
Cleanlab confident-learning · likely-wrong labels, worst first
MC-Dropout epistemic uncertainty · 20 stochastic passes
Dropout is kept active at inference and the model is run 20 times per patch. Where those passes disagree, the per-pixel standard deviation is high — the model is unsure there. High-uncertainty regions are exactly what the active-learning queue prioritises.
Confidence calibration: routing on MC-Dropout std is more trustworthy than raw softmax, which is typically over-confident on rare classes like debris.
Next-to-label queue ranked by MC-Dropout uncertainty + entropy
The ranking above is a real uncertainty computation. The retrain step needs a live GPU backend, so its projected F1 lift is a scripted illustration — no training happens in your browser.
Encoder-feature embedding · click a point to load that patch
Deepest ResNet34 encoder features per patch, reduced to 2D (—). Nearby points look alike to the model; outliers & high-uncertainty (amber-ringed) points are the most valuable to label or double-check.
Consensus labelling · inter-annotator agreement (Cohen's κ)
Three simulated annotators (a marine-ecology expert, a trained analyst, and a crowd worker) label this patch's detections. Agreement is computed live; disagreements route to adjudication.
Throughput, turnaround & auto-calibrated threshold
Active-learning routing means only low-confidence detections need a human — the rest auto-accept. Cost figures assume a $0.12/region annotation rate.
Where this graduates from a browser demo to an operational pipeline
Every box is a named open-source or managed component. The browser demo you're using exercises the Model, Uncertainty, Label QA and Curation stages on real bundled data; the ingest, orchestration and live-retrain lanes are what a full deployment adds.
Data: MARIDA (Kikaki et al. 2022, PLoS ONE) · Sentinel-2 L2A · CC BY 4.0. Model, uncertainty, label-QA & embeddings computed by GIA.
← All Labs