ssrajadh

ssrajadh / sentrysearch

Public

Semantic search over videos using Gemini Embedding 2.

chromadb dashcam gemini gemini-embedding-2 semantic-search
16
0
100% credibility
Found Mar 24, 2026 at 16 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

SentrySearch enables users to search dashcam video footage using natural language queries and automatically generates trimmed clips of the best matches.

How It Works

1
🔍 Discover SentrySearch

You stumble upon a handy tool that searches your dashcam videos by simply describing what you're looking for, like magic.

2
🛠️ Get it ready

You install it on your computer and connect a smart AI helper so it can understand video content.

3
📁 Load your videos

You show it the folder with your dashcam footage, and it quietly prepares all the clips for easy searching.

4
💭 Describe what you want

You type everyday words like 'red truck running a stop sign,' and it instantly finds the matching moments.

5
✂️ Watch the clips

It lists the best matches with times and saves trimmed video clips straight to your folder.

🎉 Find anything fast

Now you can relive key dashcam moments in seconds, without scrubbing through hours of footage.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 16 to 16 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is sentrysearch?

Sentrysearch is a Python CLI tool for semantic search over video footage, like dashcam MP4s from Tesla Sentry Mode. Feed it a directory of videos, and it indexes them into a local database using Gemini embeddings—turning hours of raw footage into searchable clips via natural language queries like "red truck running a stop sign." It auto-trims and saves the best matching clip, skipping static scenes to cut costs on Gemini's API.

Why is it gaining traction?

It stands out with direct video-to-text embedding via Gemini Embedding 2—no transcription or frame captioning needed, unlike semantic search elasticsearch or Azure pipelines. Developers dig the optimizations like preprocessing for faster/cheaper indexing and still-frame skipping, plus simple commands (sentrysearch index/search/stats) that handle ChromaDB storage under the hood. For semantic search Python or RAG apps, it's a lightweight alternative to heavier semantic search core setups.

Who should use this?

Dashcam owners sifting Tesla Sentry footage for incidents, security teams querying surveillance videos, or devs prototyping semantic search RAG with video embeddings. Ideal for hobbyists with MP4 archives or indie devs building semantic github action tools around Gemini, not enterprise semantic versioning workflows.

Verdict

Grab it if you have video search needs—docs are clear, setup is dead simple, and it works out of the box with a free Gemini key. At 16 stars and 1.0% credibility, it's early alpha (no tests, preview API), so test on non-critical footage before production.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.