Back to projects

Vision · Accessibility Jun to Sep 2022

Real-Time Object Recognition

Abstract.An assistive app that recognizes everyday objects from a live camera and announces them by audio.

Role
AI Developer Intern
Period
Jun to Sep 2022
Context
Digipodium, Inc. · Lucknow, India
Team
Solo
Table 1Results.
~1,000 Self-collected training images
Real-time Detection with audio feedback

01From problem to approach

The problem

Visually impaired users need to know what is in front of them quickly, without asking for help.

The approach

Train a detector on a self-collected dataset and speak each result aloud.

02How it works

  1. Input Camera feed Capture live video frames from the camera.
  2. Step 02 Detect (TensorFlow) A trained model identifies objects in each frame.
  3. Step 03 Annotate (OpenCV) Draw bounding boxes and labels with confidence scores.
  4. Output Announce by audio Speak the detected objects aloud.
Fig. 1Real-Time Object Recognition, step by step.

03What I built

  1. Collected and labelled a dataset of about 1,000 images and trained a TensorFlow detection model.
  2. Built the OpenCV real-time video pipeline and a Streamlit web UI with audio feedback.
  3. Designed the database and wrote the basic design documents (ER diagrams, functional and screen design).
Live detection of a bed and a chair with confidence scores
Fig. 2Live detection of a bed and a chair with confidence scores

04Technologies

  • Python
  • TensorFlow
  • OpenCV
  • Pandas
  • Streamlit
  • SQL Server

On request

Request my resume

I share my resume on request. Tell me a little about yourself, and I'll email it to you once I've reviewed your request.

I am a…

Which document?

0 / 100 words

Your details are used only to handle this request and are never shared.