Families In the Wild

A Kinship Recognition Benchmark

Families In the Wild (FIW) is the largest and most comprehensive image database for automatic kinship recognition. Our motivation is to provide the resource needed for kinship recognition technologies to transition from research-to-reality. with over 11,932 family photos of 1,000 families FIW closely reflects the true data distribution of families worldwide (see Database for more information). There are many directions for FIW to take throughout the machine vision and related research communities (e.g., in relation to benchmarks (see Challenges and Results for details), new benchmarks, generative models, multi-modal learning, to name a few). In terms of its practical value, many could benefit from FIW as well, such as the consumer (e.g. automatic photo library management), scholar (e.g. historic lineage & genealogical studies), analyzer (e.g. social-media-based analysis), investigator (e.g. missing persons and human traffickers).

Balanced Faces in the Wild (BFW)

A Face Recognition Benchmark

Intended to address problems of bias in facial recognition, we built BFW as a labeled data resource made available for evaluating recognitiion systems on a corpus of facial imagery made-up of EQUAL face count for all subjects, which are EQUAL across demographics, and, thus, face data balanced in faces per subject, subjeccts per ethniciity, ethnicity (or faces) per gender.

EV-Action: Electromyography-Vision Multi-Modal Action Dataset

Multi-modal human action analysis is a critical and attractive research topic. However, the majority of the existing datasets only provide visual modalities (i.e., RGB, depth and skeleton). To make up this, we introduce a new, large-scale EV-Action dataset in this work, which consists of RGB, depth, electromyography (EMG), and two skeleton modalities. Compared with the conventional datasets, EV-Action dataset has two major improvements: (1) we deploy a motion capturing system to obtain high quality skeleton modality, which provides more comprehensive motion information including skeleton, trajectory, acceleration with higher accuracy, sampling frequency, and more skeleton markers. (2) we introduce an EMG modality which is usually used as an effective indicator in the biomechanics area, also it has yet to be well explored in motion related research. To the best of our knowledge, this is the first action dataset with EMG modality. We hope this dataset can make significant contributions to human motion analysis, computer vision, machine learning, biomechanics, and other interdisciplinary fields.