Computer Vision

The BD-DET is obtained by professional industrial cameras equipped with a ring light source, with a resolution of 2456×2048. The basic structure of the bearing includes inner and outer rings, a cage, and a sealing cover. Defects are mainly concentrated in the sealing cover area, and are divided into two types: scratches and presses. During the data acquisition process, some data only contain one type of defect, either scratches or presses, while the remaining data contain a mixture of the two defects.


In this investigation, the researchers have used a commercially available millimeter-wave (MMW) radar to collect data and assess the performance of deep learning algorithms in distinguishing different objects. The research looks at how varied ambiance factors, such as height, distance, and lighting, affect object recognition ability in both static and dynamic stages of the radar.


Three exemplary knee bone models derived from ultrasound imaging and their respective magnetic resonance imaging reference.

Apart from the ground truth, a partial scan as accessable by ultrasound imaging as well as full bone model computed by a statistical shape model is provided.


The Extended Moulouya Bird Detection Dataset (E-Moulouya BDD) is a comprehensive collection of annotated images proposed for bird detection. The dataset is a combination of three datasets, namely the XMBA dataset, the Rest Birds Dataset, and the D-Birds Dataset, which were merged and cleaned up to provide a consistent and unified labeling format. The E-Moulouya BDD comprises 13,000 annotated images, labeled with a standardized format to ensure consistency across the dataset. The dataset is publicly available on IEEE DATAPORT, making it an ideal resource for use by the scientific community.


The dataset consists of NumPy arrays for each alphabet in Indian Sign Language, excluding 'R'. The NumPy arrays denote the (x,y,z) coordinates of the skeletal points of the left and right hand (21 skeletal points each) for each alphabet. Each alphabet has 120 sequences, split into 30 frames each, giving 3600 .np files per alphabet, using MediaPipe.

The dataset is created on the basis of skeletal-point action recognition and key-point collection.


When fuel materials for high-temperature gas-cooled nuclear reactors are quantification tested, significant analysis is required to establish their stability under various proposed accident scenarios, as well as to assess degradation over time. Typically, samples are examined by lab assistants trained to capture micrograph images used to analyze the degradation of a material. Analysis of these micrographs still require manual intervention which is time-consuming and can introduce human-error.


The Dataset consists of two videos, one recorded with blindfold on and the other without blindfold recorded using a 1080p Intel RealSense depth camera. It contains the videos, images extracted using ffmpeg and processed video which is made of a video with skipped frames created using ffmpeg. The scope of the dataset is for machine vision purposes to allow for tasks such as instance segmentation. A hat fixed on the head of a blindfolded person is used to record walking activities.


Human activity recognition, which involves recognizing human activities from sensor data, has drawn a lot of interest from researchers and practitioners as a result of the advent of smart homes, smart cities, and smart systems. Existing studies on activity recognition mostly concentrate on coarse-grained activities like walking and jumping, while fine-grained activities like eating and drinking are understudied because it is more difficult to recognize fine-grained activities than coarse-grained ones.