Computer Vision Toolbox

Design and test computer vision, 3D vision, and video processing systems

Computer Vision Toolbox provides algorithms, functions, and apps for designing and testing computer vision, 3D vision, and video processing systems. You can perform object detection and tracking, as well as feature detection, extraction, and matching. You can automate calibration workflows for single, stereo, and fisheye cameras. For 3D vision, the toolbox supports visual and point cloud SLAM, stereo vision, structure from motion, and point cloud processing. Computer vision apps automate ground truth labeling and camera calibration workflows.

You can train custom object detectors using deep learning and machine learning algorithms such as YOLO, SSD, and ACF. For semantic and instance segmentation, you can use deep learning algorithms such as U-Net and Mask R-CNN. The toolbox provides object detection and segmentation algorithms for analyzing images that are too large to fit into memory. Pretrained models let you detect faces, pedestrians, and other common objects.

You can accelerate your algorithms by running them on multicore processors and GPUs. Toolbox algorithms support C/C++ code generation for integrating with existing code, desktop prototyping, and embedded vision system deployment.

What Is Computer Vision Toolbox?

Image and Video Ground Truth Labeling

Automate labeling for object detection, semantic segmentation, instance segmentation, and scene classification using the Video Labeler and Image Labeler apps.

Documentation | Examples

Pedestrians, cars, and buses labeled using instance segmentation.

Deep Learning and Machine Learning

Train or use pretrained deep learning and machine learning based object detection and segmentation networks. Evaluate the performance of these networks and deploy them using C/C++ or CUDA^® code.

Documentation | Examples

Original pill image and the same image with anomaly markings.

Automated Visual Inspection

Use the Automated Visual Inspection Library in Computer Vision Toolbox to identify anomalies or defects to assist and improve quality assurance processes in manufacturing.

Documentation | Examples

Camera Calibration

Estimate the intrinsic, extrinsic, and lens-distortion parameters of monocular and stereo cameras using the camera calibration and stereo camera calibration apps.

Documentation | Examples

A dense reconstruction of a scene applying visual SLAM to data from an RGB-D camera.

Visual SLAM and 3D Vision

Extract the 3D structure of a scene from multiple 2D views. Estimate camera position and orientation with respect to its surroundings. Refine pose estimates using bundle adjustment and pose graph optimization.

Documentation | Examples

Lidar and 3D Point Cloud Processing

Segment, cluster, downsample, denoise, register, and fit geometrical shapes with lidar or 3D point cloud data. Lidar Toolbox provides additional functionality to design, analyze, and test lidar processing systems.

Documentation | Examples

Detecting a box in a scene using point feature matching.

Feature Detection, Extraction, and Matching

Detect, extract, and match features such as blobs, edges, and corners, across multiple images. Features matched across images can be used for registration, object classification, or in complex workflows such as SLAM.

Documentation | Examples

Detecting multiple pedestrians in the area of interest in a car dashcam video.

Object Tracking and Motion Estimation

Estimate motion and track objects in video and image sequences.

Documentation | Examples

Code Generation and Third Party Support

Use the toolbox for rapid prototyping, deploying, and verifying computer vision algorithms. Integrate OpenCV-based projects and functions into MATLAB and Simulink.

Documentation | Examples

Product Resources:

Documentation Examples Videos Functions Blocks Hardware support Requirements Release notes

Drass Develops Deep Learning System for Real-Time Object Detection in Maritime Environments

“From data annotation to choosing, training, testing, and fine-tuning our deep learning model, MATLAB had all the tools we needed—and GPU Coder enabled us to rapidly deploy to our NVIDIA GPUs even though we had limited GPU experience.”
Valerio Imbriolo, Drass Group

View more customer stories