connecting…
Computer-Vision Workbench

See what the model sees,
|

Lumen runs pretrained vision models on your images and shows their reasoning — attention heatmaps, confidence, and labels. One workbench, many models: skin-lesion analysis, object detection, and more to come.

2
Models live
Grad-CAM
Explainability
Real-time
Inference
Capabilities

What Lumen can do

Every model in the workbench doesn't just answer — it shows its working. Here's what you get on each image.

Medical

Skin Lesion Analysis

Classifies dermatoscopic images as benign or malignant using a fine-tuned ResNet-50, with the confidence behind every call.

Detection

Object Explorer

Point the YOLOv8 detector at any photo and get labeled bounding boxes across 80 everyday object classes in real time.

Explainability

Attention Heatmaps

Grad-CAM overlays reveal exactly which pixels drove the model's decision — so you see where it looked, not just what it said.

Honesty

Calibrated Confidence

Probabilities for each class, shown honestly. A borderline call looks borderline — no false certainty dressed up as a verdict.

Speed

Real-Time Inference

Models run on a local FastAPI backend, returning predictions and visual explanations within seconds of upload.

Architecture

Extensible by Design

One workbench, many models. New vision tasks slot in as additional tabs — the interface grows as the project does.

Skin Lesion Analysis

Classifies a dermatoscopic image as benign or malignant and reveals where the model focused.
ResNet-50 · Grad-CAM
Demonstration only — not a medical diagnosis. This tool illustrates how a vision model attends to an image. It cannot diagnose disease. Anyone with a skin concern should consult a qualified dermatologist.
Input image
Drop a lesion image or click to browse
JPG · PNG · WEBP
Model readout
Awaiting input
Upload a lesion image to see the classification, confidence, and attention heatmap.