Welcome to our Hackathon-2022. Through this hackathon we are trying to select interns have passion to work in computer vision and deep learning. We have internships positions for a couple of projects from Defect Recognition to Multi-Camera Multi object Tracking. You can select either C++/Python to solve these coding assignments. Best of luck everyone
- Use issues to ask any queries
- Write an algorithm that will detect the defect in the images given below. Breakup of the mark is as given below
Flowchart - 20 marks
Basic algorithm to find defects - 40 marks
Localizing the defect - 20
Classify the defect to flashes and cut marks - 20
- Flashes
- Cuts
- Write a program to plot following equation over time(that is values of x,y,z). Assume
$x,y,z$ are points in 3d space in which$x_{0} = 0, y_{0} = 1, z_{0} = 1.05$ and$a = 10,b = 28,c = 2.667$ are the parameters and$\dot{x} = \frac{dx}{dt}$ , similar for$y$ and$z$
$$\dot{y} = bx - y - xz$$
$$\dot{z} = xy - cz$$
Tip : This is an iterative equation. Try to find the points over time by writing derivative equation
Total : 100 marks
- The objective is to predict the location and type of vehicle found in the scene using the images provided. The images have annotations in xml format, and the task is to create an Object Detector and Classifier.Eg: Cars, Bus, Motorbikes etc. Select 3 or 4 classes from the provided dataset and build the training module. Each Image may or may not have the classes to identify. You can find description along with the dataset.
Dataset for : https://www.kaggle.com/datasets/pratikbarua/vehicle-detection-dataset
Visualize the object data distribution as graph or in any meaningful representation. Eg. number of objects in each classes, Area, width and height etc.- 10 marks
Flowchart/Blockdiagram - 25 marks
Create an algorithm to identify the objects and classify - 40 (Percentile. The person who submitting the highest accuracy will get 40 and others mark will be normalised)
Evaluation metrics and plots - 25
- Late submissions will not be considered for evaluation
- The submission consists of a document which describes the flowcharts and explain why you make use of the particular algorithm to solve the problem along with the results and discussions on the accuracy (if your model is not working on some data, show the same and explain why it is not working as per your understanding)
- The code has to be properly commented and should follow PEP8 rules. The code should be written in modular format. For example, data preparation, model training and model inference and result visualisations