We are creating an Object Recognition Program which will be able to detect objects and accurately classify said objects in the given image file or video file with correct labels. We will implement neural networks in our project. Our project aims to utilize the full potential of geometric cues and exploit them to extract other features in a robust, computationally efficient manner. With each assessment, we will provide a more refined version of our final objective, incorporating as many of the desired features as possible in the time given.
Object Recognition is a computer technology related to computer vision and image processing that deals with recognising instances of semantic objects of a certain class.
A semantic object is a representation of a collection of attributes that describe an identifiable thing in the user environment
Deep learning is a broader family of machine learning, which uses data representations to learn, rather than task-based algorithms. Most modern deep learning models are based on an Artificial Neural Network, which uses a cascade of multiple layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input.
Python
Tensorflow (A Python library for fast numerical computing created and released by Google)
OpenCV (Open Source Computer Vision is a library of programming functions mainly aimed at real-time computer vision)
Image AI API (A python library built with self-contained Computer Vision capabilities)
PHP
CSS (Bootstrap)
Javascript/HTML
TensorFlow is used for high performance numerical computation and comes with strong support for deep learning and the flexible numerical computation core is used across many other scientific domains. We will use TensorFlow in our project for training our program and implementing our model.
OpenCV (Open Source Computer Vision Library) is an open source computer vision and machine learning software library. OpenCV was built to provide a common infrastructure for computer vision applications and to accelerate the use of machine perception. It will be used in our project for real-time image processing.
We will use technologies such as Deep Learning and Image processing. We will also try to understand the concept of our project better using methodologies such as Object Detection, Object Segmentation, Image Segmentation and Semantic Image Segmentation.
