Case study
AS-One: Unified Computer Vision Framework for Rapid AI Development
- Client: Augmented Startups
- Industry: Open Source AI Infrastructure
- AI capability: Computer Vision
- GitHub stars on AS-One
- 580+
- Detector versions behind one API
- YOLOv5–v9
The challenge
Every detector and tracker speaks its own dialect
Developers building computer vision applications often struggle with incompatible libraries for detection, tracking, and inference. Every framework requires different implementations and dependencies, leading to complex integration pipelines, redundant code, and high development overhead for experimentation. Augmented Startups wanted a solution that allowed for seamless 'plug-and-play' functionality across various YOLO variants and popular trackers like ByteTrack and DeepSORT.
What we built
One Python API over many models
Swap the detector or the tracker without rewriting the pipeline.
Detector
- YOLOv5
- YOLOv6
- YOLOv7
- YOLOv8
- YOLOv9
Tracker
- ByteTrack
- DeepSORT
One
AS-One API
1/4Choose a detector. YOLOv5 through v9 sit behind the same call, so the rest of your code does not change.
Every detector and tracker ships its own interface, so swapping one means rewriting the pipeline. AS-One puts them behind a single Python API: pick a detector, pick a tracker, run the same code.
Modular by design
Inference is decoupled from tracking, so either side can be swapped without touching the other.Standardised API
One call signature covers YOLOv5 to YOLOv9 and the trackers behind them.ByteTrackDeepSORTMultiple runtimes
PyTorch, ONNX and CoreML builds keep deployment platform-agnostic.Open source
Used by developers worldwide, with 580+ GitHub stars and a setup tutorial.
The demo app, on two kinds of footage
Cars tracked and counted as they cross a line, with a confidence slider and class filter. 
The same app filtered to backpacks and handbags in a crowd.
The impact
An open-source tool developers actually use
- 580+ GitHub stars
- YOLOv5 to v9 behind one API
- PyTorch, ONNX and CoreML runtimes
AS-One has gained 580+ GitHub stars and is widely used by developers and researchers worldwide. It has significantly reduced integration complexity, simplified AI experimentation, and become a powerful open-source tool for rapidly prototyping production-ready vision applications.
Keep exploring
Related case studies
Real Estate & PropTechComputer VisionAI Property Image Intelligence for Listing Optimization & Compliance
PropTech founders and real estate leaders understand that property photos are the most critical asset in closing deals. PropTexx needed an automated way to enforce strict listing compliance and extract valuable insights from millions of unstructured real estate images without slow manual review.
- Property features tagged automatically
- 50+
- Compliance violation types detected
- 15+
Retail & E-commerceComputer VisionSneaker Visual Retrieval System
AlwaysLegit needed a way for their team at expos to instantly identify shoes from photos and match them against a massive 150K+ SKU database, often in challenging real-world lighting and backgrounds.
- SKUs searchable from a single photo
- 150K+
- Retrieval time per query
- <1s
Sports & MediaComputer VisionSoccer Match Analytics & Insights
The aim of this project is to give soccer analytics of matches being played. The goal is to build an AI tool that takes as input a video of a soccer match and then analyzes each player and extracts different analytics. This included: 1. Player and ball detection, 2. Player and ball tracking, 3. Player re-identification, 4. Field detection, 5. Event detection including passes, intercepts, goals etc.
Multi-stage CV pipeline. Real-time event detection. Player re-identification.
Read the case study
Book a strategy session
Talk to an AI engineer about your project
Tell us what you want to automate. The first call is a 30-minute working session with an engineer, not a sales pitch.
- Send the form, it takes 2 minutes
- We reply within 1 business day, under NDA if you need it
- A 30-minute call to scope feasibility and next steps