What are the best AR computer vision portfolio projects to showcase your skills?

1 Face filters

Face filters are one of the most popular and fun features of AR apps like Snapchat and Instagram. They use computer vision to detect and track facial features, and apply digital effects such as masks, stickers, or animations. To create your own face filters, you can use tools like Spark AR Studio, Lens Studio, or OpenCV. You can also experiment with different types of filters, such as emotion recognition, face swapping, or beauty enhancement. Face filters are a great way to showcase your skills in face detection, facial landmark extraction, and image processing.

2 Object recognition

Object recognition is the ability to identify and classify objects in an image or video. It is a key component of many AR applications, such as product scanning, navigation, or education. For example, you can create an AR app that recognizes different types of plants or animals, and provides information or feedback to the user. To build your own object recognition project, you can use frameworks like ARKit, ARCore, or Vuforia, and libraries like TensorFlow, PyTorch, or Keras. You can also use pre-trained models or datasets, or create your own custom ones. Object recognition is a great way to showcase your skills in deep learning, computer vision, and data analysis.

3 Image segmentation

Image segmentation is the process of dividing an image into multiple regions based on some criteria, such as color, texture, or shape. It is useful for many AR applications, such as background removal, virtual try-on, or photo editing. For example, you can create an AR app that segments the user's hair or clothes, and allows them to change their style or color. To implement your own image segmentation project, you can use techniques like thresholding, clustering, or convolutional neural networks. You can also use tools like OpenCV, scikit-image, or Mask R-CNN. Image segmentation is a great way to showcase your skills in image processing, machine learning, and computer graphics.

4 Marker-based tracking

Marker-based tracking is a technique that uses predefined markers, such as QR codes, barcodes, or logos, to anchor digital content to the real world. It is widely used for AR applications, such as advertising, gaming, or art. For example, you can create an AR app that displays a 3D model or animation when the user scans a marker with their camera. To develop your own marker-based tracking project, you can use tools like ARKit, ARCore, or Vuforia, and frameworks like Unity, Unreal Engine, or SceneKit. You can also design your own markers or use existing ones. Marker-based tracking is a great way to showcase your skills in computer vision, 3D modeling, and game development.

5 Gesture recognition

Gesture recognition is the ability to recognize and interpret human gestures, such as hand movements, facial expressions, or body poses. It is useful for AR applications that require natural and intuitive interaction, such as gaming, fitness, or entertainment. For example, you can create an AR app that lets the user control a virtual character or object with their gestures. To build your own gesture recognition project, you can use tools like OpenCV, MediaPipe, or Vision, and libraries like TensorFlow, PyTorch, or PoseNet. You can also use sensors like Kinect, Leap Motion, or ARKit's TrueDepth camera. Gesture recognition is a great way to showcase your skills in human-computer interaction, computer vision, and machine learning.

6 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?