--- title: Tracer Vision — Offline Image-Similarity Discovery for ICAC description: Tracer Vision by Kire Labs is a free, offline image- and video-similarity discovery tool for ICAC task forces. Find visually similar material with local SigLIP embeddings and export Project VIC 2.0 for Griffeye. url: https://kirelabs.com/tracer-vision --- # Tracer Vision **See what looks alike. Not just what matches exactly.** Tracer Vision compares unknown images and videos against a reference corpus you control—using a local visual-similarity model (SigLIP) to surface material that hash matching misses. It runs fully offline and exports Project VIC 2.0 for Griffeye. - **Offline** — matching runs with zero network access - **Free** — for U.S. law enforcement & ICAC task forces - **SigLIP** — local perceptual image + video matching - **Project VIC** — 2.0 JSON export for Griffeye GID ## The problem Hash hits are easy. Everything else is the hard part. Known-file hashing catches what's already catalogued. But altered, re-cropped, re-compressed, and never-before-seen material slips straight through—and your analysts have no fast way to know which of thousands of unknowns to look at first. 1. **Hash matching only catches exact copies.** MD5 and SHA-1 are bit-for-bit. Change a single pixel, re-encode, or re-crop and the hash changes—so the file reads as brand new even when it's the same scene you've seen a hundred times. 2. **Triage is manual and slow.** Faced with a folder of unknowns, analysts have no signal for where to start. Reviewing everything by eye is slow and keeps people in front of harmful material longer than necessary. 3. **Cloud tools send sensitive media off-host.** Many similarity tools require uploading the very material you're investigating. For this work, sending it to someone else's server is a non-starter. ## Features Every check run gives each unknown up to three independent signals—visual similarity, hash matches, and review priority. You read them together; none overrides the others. - **SigLIP visual similarity** — The primary signal. Each unknown is embedded locally with Google's SigLIP encoder and compared to your seed corpus by squared-Euclidean distance. Perceptual matching catches variants, crops, and re-compressions that hashing never will—reported as a tunable similarity band. - **Project VIC hash matching** — Import one or more Project VIC OData datasets and every unknown's MD5 and SHA-1 is looked up against them. Hits appear as chips next to the SigLIP score—they never change or override the visual-similarity result. - **Image & video** — Videos are sampled frame-by-frame with ffmpeg, each frame embedded, and the closest-matching frame reported as the finding. Per-frame embeddings are cached, so re-checking the same video later never needs ffmpeg again. - **Project VIC 2.0 export** — One click produces a Project VIC 2.0 JSON report (or CSV) ready to ingest into Griffeye GID and your existing case management. Tracer Vision feeds your toolchain instead of replacing it. - **Offline-first & private** — A single binary that runs on a laptop—no server, no account, no per-image cloud calls. No telemetry, no auto-update beacon. Seeds are never rendered; unknowns stay masked behind a click-to-reveal so analysts opt in to view each file. - **Content Safety priority** — Optional and opt-in per run: with an API key, each unknown can be classified by Google's Content Safety endpoint into a review priority (VERY_LOW…VERY_HIGH). Only the embedding vector is sent—never the image. The one part of the workflow that touches the network, and it's off by default. ## How it works Seed in, shortlist out. Tracer Vision sits at the discovery step: curate what you're looking for, point it at a pile of unknowns, and it tells you which items most resemble your reference set—so analysts review the right things first. 1. **Load a seed corpus.** Point it at folders of reference images. Tracer Vision computes and stores embeddings and hashes—no source paths, no thumbnails. The originals can leave the host afterward. 2. **Check unknowns.** Drop a folder of unknown images and videos. Each is embedded locally and compared to every seed; the closest match is reported with a similarity band, VIC hits, and an optional review priority. 3. **Read the signals.** A results table ranks unknowns by similarity, with hash chips and priority side by side. History persists every check so you can re-match against an updated corpus or find similar unknowns. 4. **Export to Griffeye.** Produce a Project VIC 2.0 JSON (or CSV) report and ingest it into Griffeye GID. The findings flow into the toolchain and case management your team already runs. ## A desktop app and a CLI Most analysts live in the desktop GUI—Seeds, Check, History, and Settings views with drag-and-drop, masked previews, and one-click export. For labs that script their pipeline, the same engine drives a full CLI. Both produce identical Project VIC output. - Windows (primary), Linux (supported), macOS (best-effort) - Single binary—no server, no database to administer - Tunable concurrency for fast batch runs on a workstation ```sh # Build a reference corpus $ tracer-vision ingest --dir ./seeds --kind seed # Check a folder of unknowns, write a report $ tracer-vision match --dir ./unknowns --out report.json # Load a Project VIC dataset (additive hash signal) $ tracer-vision vic-import --file vic.json --source NCMEC # …or just open the desktop app $ tracer-vision gui ``` ## Scope: a discovery aid, not a system of record Tracer Vision points your analyst at the items most likely to be worth a closer look first. It runs on material you already possess. Every signal it reports—visual similarity, hash matches, review priority—is a hint; an analyst always makes the call. **What it is** - A fast triage and discovery aid for previously-unseen material - Offline-first, single-user, runs entirely under your control - A feeder for Griffeye and your existing case management **What it isn't** - Not evidence collection—no chain-of-custody claims - Not a cloud service—your media never leaves the host - Not an automatic decision-maker—findings always need review ## Get access Free for qualifying agencies. Tracer Vision is licensed free of charge to qualifying U.S. law enforcement—solely for investigating, deterring, or prosecuting crimes against children. Tell us about your agency and we'll get you set up. - Free for U.S. law enforcement under the Tracer Vision EULA - Offline by default—no data leaves your host - Built in cooperation with Wyoming ICAC (and Google CSAM team) Request access with the form at . The form is protected by a Cloudflare Turnstile challenge, so it must be completed by a person in a browser. --- Tracer Vision by [Kire Labs, LLC](https://kirelabs.com/) · © 2026 Kire Labs, LLC. All rights reserved.