Given screenshot shows how AI can run forensics on just a sketch of the suspect

Inspiration

Every year, thousands of cold cases pile up in archives—boxes filled with fading witness sketches, grainy CCTV footage, and dusty autopsy reports. The problem isn't a lack of evidence; it's that the evidence is fragmented. A human detective has to manually cross-reference a witness saying 'blue eyes' with a blurry photo showing 'light irises' and a report mentioning 'mid-30s'. It's cognitive overload.

What it does !

We watched a documentary on how forensic artists take weeks to reconstruct a face from notes. We realized: What if we could give every detective a digital forensic artist that works in seconds? That’s why we built Chrono-Sculpt. We wanted to use Multimodal AI to stitch these scattered signals into a single, high-fidelity lead

Future-Scope :

1) 3D Volumetric Rendering: Next, we will pipe the output into a 3D engine to generate a rotatable head model for VR line-ups. 2) Audio-to-Face: Using the audio files to predict approximate skull structure based on voice resonance (a real forensic technique). 3) Cross-Jurisdiction Linking: Automatically matching our generated profile against missing persons databases in other states.