Introducing the Roboflow Deployment Tab

Roboflow · Beginner ·👁️ Computer Vision ·4y ago

Key Takeaways

The Roboflow Deployment Tab is introduced, allowing users to easily deploy and use their computer vision models after training, with features such as an inference widget, documentation, and code snippets for hosted inference API and edge deployment.

Full Transcript

hey everybody it's Lucas from roboflow I'm here to show you guys the new deploy page that we made so it's as easy as possible start using your computer vision models as soon as they finish training we did this by making a new inference widget that accepts your webcam video and image formats in the platform as well as providing documentation and code Snippets formatted with the model endpoint version and your API key so you can copy and paste them and put them directly into your application and get off the ground running so I'm going to show you guys today uh using a simference widget and some of the code Snippets so here I'm going to start by choosing from any of the trained versions I'm just going to leave it as the default which is the latest and then over here you can see the training results and the stats as well as some nice visuals and then here with the inference widget I can choose an image from the test set and I can change the confidence threshold over here and the overlap threshold and I can also use my webcam here you go and you can use these thresholds as well with the videos and I can also paste a URL to an image so here just a quick Google search take this URL drop it here or I can also just drag this over and do that and I can also upload images or you can upload them directly if you don't want to drag and drop them and then I'm going to show you with a video I'm just going to drop this in here and I'm going to switch this rabbit model and then one of the nice things here is this video is in slow motion but I can change the speed so I'm going to put it at two times speed so it's a little bit more normal and then at the bottom here these are the code symptoms that I was talking about so you can either use our hosted inference API for auto scale deployment or you can deploy to the edge with one of these other deployment options and as I said these are all formatted they have uh the model endpoint the version and your API keys that are already formatted so you can just copy and paste and you're on your way and then lastly we have over here some resources on the right so here's our full documentation so you can find out more about inference with roboflow happy inferencing

Original Description

We recently released a new deployment tab to the Roboflow Platform (app.roboflow.com) and Roboflow Universe (universe.roboflow.com) so you can easily use your computer vision models after they have been trained.
Watch on YouTube ↗ (saves to browser)
Sign in to unlock AI tutor explanation · ⚡30

Playlist

Uploads from Roboflow · Roboflow · 0 of 60

← Previous Next →
1 YOLOv3 PyTorch Notebook Tutorial
YOLOv3 PyTorch Notebook Tutorial
Roboflow
2 How to Train YOLOv4 on a Custom Dataset (PyTorch)
How to Train YOLOv4 on a Custom Dataset (PyTorch)
Roboflow
3 How to Train YOLOv5 on a Custom Dataset
How to Train YOLOv5 on a Custom Dataset
Roboflow
4 How to Use the Roboflow Dataset Health Check
How to Use the Roboflow Dataset Health Check
Roboflow
5 What is Mean Average Precision (mAP)?
What is Mean Average Precision (mAP)?
Roboflow
6 How to Use the Roboflow Model Library
How to Use the Roboflow Model Library
Roboflow
7 How to Train EfficientDet in TensorFlow 2 Object Detection
How to Train EfficientDet in TensorFlow 2 Object Detection
Roboflow
8 How to Train YOLO v4 Tiny (Darknet) on a Custom Dataset
How to Train YOLO v4 Tiny (Darknet) on a Custom Dataset
Roboflow
9 Ask the Roboflow Team Anything - Episode 1
Ask the Roboflow Team Anything - Episode 1
Roboflow
10 Exploring The COCO Dataset
Exploring The COCO Dataset
Roboflow
11 Community Spotlight: Improving Uno with Computer Vision
Community Spotlight: Improving Uno with Computer Vision
Roboflow
12 Mosaic Data Augmentation - Deep Dive
Mosaic Data Augmentation - Deep Dive
Roboflow
13 Hands on with the OAK-1
Hands on with the OAK-1
Roboflow
14 Glenn Jocher: What is New in YOLO v5?
Glenn Jocher: What is New in YOLO v5?
Roboflow
15 How to Use Amazon Rekognition Custom Labels and Roboflow to Build an Object Detection Model
How to Use Amazon Rekognition Custom Labels and Roboflow to Build an Object Detection Model
Roboflow
16 An Interview with Brandon Gilles, Luxonis Founder and OAK Chief Architect
An Interview with Brandon Gilles, Luxonis Founder and OAK Chief Architect
Roboflow
17 How to Train a Custom Mobile Object Detection Model (with YOLOv4 Tiny and TensorFlow Lite)
How to Train a Custom Mobile Object Detection Model (with YOLOv4 Tiny and TensorFlow Lite)
Roboflow
18 Tackling the Small Object Problem in Object Detection
Tackling the Small Object Problem in Object Detection
Roboflow
19 Fast.ai v2 Released - What's New?
Fast.ai v2 Released - What's New?
Roboflow
20 Teaser: Roboflow Train (1-Click Computer Vision AutoML)
Teaser: Roboflow Train (1-Click Computer Vision AutoML)
Roboflow
21 How to Train a Custom Resnet34 Image Classification Model
How to Train a Custom Resnet34 Image Classification Model
Roboflow
22 How to Label Images for Object Detection with CVAT
How to Label Images for Object Detection with CVAT
Roboflow
23 Deploy YOLOv5 to Jetson Xavier NX at 30 FPS
Deploy YOLOv5 to Jetson Xavier NX at 30 FPS
Roboflow
24 Elisha Odemakinde Hosts Roboflow ML Engineer, Jacob Solawetz
Elisha Odemakinde Hosts Roboflow ML Engineer, Jacob Solawetz
Roboflow
25 Getting Started with VoTT - Computer Vision Annotation
Getting Started with VoTT - Computer Vision Annotation
Roboflow
26 How to Manage Classes in Object Detection (Rename, Combine, Balance)
How to Manage Classes in Object Detection (Rename, Combine, Balance)
Roboflow
27 How to Train YOLOv4 on a Custom Dataset in Darknet
How to Train YOLOv4 on a Custom Dataset in Darknet
Roboflow
28 Is Grayscale a Preprocessing or Augmentation Step in Computer Vision?
Is Grayscale a Preprocessing or Augmentation Step in Computer Vision?
Roboflow
29 Getting Started with Image Data Augmentation
Getting Started with Image Data Augmentation
Roboflow
30 Glenn Jocher: Image Augmentation in YOLO v5 and Beyond
Glenn Jocher: Image Augmentation in YOLO v5 and Beyond
Roboflow
31 GA Hosts Roboflow - Healthcare and AI
GA Hosts Roboflow - Healthcare and AI
Roboflow
32 How do self driving cars know when to stop?
How do self driving cars know when to stop?
Roboflow
33 What is PASCAL VOC XML?
What is PASCAL VOC XML?
Roboflow
34 AutoML Showdown: Google vs Amazon vs Microsoft
AutoML Showdown: Google vs Amazon vs Microsoft
Roboflow
35 How is computer vision changing manufacturing?
How is computer vision changing manufacturing?
Roboflow
36 The Alphabet in American Sign Language
The Alphabet in American Sign Language
Roboflow
37 Luxonis OAK-D: Computer Vision on Device
Luxonis OAK-D: Computer Vision on Device
Roboflow
38 How to Train a Custom Faster R-CNN Model with Facebook AI's Detectron2 | Use Your Own Dataset
How to Train a Custom Faster R-CNN Model with Facebook AI's Detectron2 | Use Your Own Dataset
Roboflow
39 TensorFlow vs PyTorch: Fireside
TensorFlow vs PyTorch: Fireside
Roboflow
40 Occlusion Techniques in Computer Vision
Occlusion Techniques in Computer Vision
Roboflow
41 A Customizable Web Application for Your Computer Vision Model
A Customizable Web Application for Your Computer Vision Model
Roboflow
42 Model Tradeoffs and the Future of Computer Vision
Model Tradeoffs and the Future of Computer Vision
Roboflow
43 Designing an Augmented Reality Board Game App
Designing an Augmented Reality Board Game App
Roboflow
44 YOLOv4 - Advanced Tactics
YOLOv4 - Advanced Tactics
Roboflow
45 How to Use CreateML and Build a Computer Vision iPhone App | AR Object Detection
How to Use CreateML and Build a Computer Vision iPhone App | AR Object Detection
Roboflow
46 Fireside Chat: Computer Vision in Agriculture
Fireside Chat: Computer Vision in Agriculture
Roboflow
47 Scaled-YOLOv4 Tops EfficientDet: Research Rundown
Scaled-YOLOv4 Tops EfficientDet: Research Rundown
Roboflow
48 What is Image Preprocessing?
What is Image Preprocessing?
Roboflow
49 Building a Community of Creators with BlkArthouse and Von Deon
Building a Community of Creators with BlkArthouse and Von Deon
Roboflow
50 How to Train Scaled-YOLOv4 to Detect Custom Objects
How to Train Scaled-YOLOv4 to Detect Custom Objects
Roboflow
51 Intro to Computer Vision: Fireside
Intro to Computer Vision: Fireside
Roboflow
52 The Best Way to Annotate Images for Object Detection
The Best Way to Annotate Images for Object Detection
Roboflow
53 The Computer Vision Process: Fireside
The Computer Vision Process: Fireside
Roboflow
54 How to Annotate Images with Your Team Using Roboflow
How to Annotate Images with Your Team Using Roboflow
Roboflow
55 Introducing the Roboflow Object Count Histogram
Introducing the Roboflow Object Count Histogram
Roboflow
56 How Fast is the M1 at Machine Learning? Benchmarking Apple's M1 and Intel's Chips
How Fast is the M1 at Machine Learning? Benchmarking Apple's M1 and Intel's Chips
Roboflow
57 CLIP: OpenAI's amazing new zero-shot image classifier
CLIP: OpenAI's amazing new zero-shot image classifier
Roboflow
58 How I hacked my Nest camera to run custom models
How I hacked my Nest camera to run custom models
Roboflow
59 Getting Started with the Roboflow Inference API
Getting Started with the Roboflow Inference API
Roboflow
60 Transfer Learning in Computer Vision | What, How, Why
Transfer Learning in Computer Vision | What, How, Why
Roboflow

The Roboflow Deployment Tab makes it easy to deploy and use computer vision models after training, with features such as an inference widget and code snippets for hosted inference API and edge deployment. This allows users to quickly integrate their models with applications and start using them in real-world scenarios.

Key Takeaways
  1. Choose a trained model version
  2. Use the inference widget to test the model with images or videos
  3. Adjust confidence and overlap thresholds
  4. Use the hosted inference API or deploy to the edge
  5. Copy and paste code snippets into your application
💡 The Roboflow Deployment Tab provides a streamlined way to deploy and use computer vision models, making it easier for users to integrate their models with applications and start using them in real-world scenarios.

Related Reads

Up next
9-Phase Computer Vision Roadmap 2026 | AI & Deep Learning | #shorts
SCALER
Watch →