A benchmark for gait impairment research
The last decade has witnessed a growing interest in clinical applications of gait assessment such as rehabilitation, medical diagnosis, and detection of medical emergencies in hospital environments. Recent studies have focused directly on describing and detecting gait impairment. The aim of this database is to provide a benchmark to allow the definition of measures of gait regularity and validate their effectiveness over abnormal gait sequences.
The INIT Gait Database consists of sequences of high-quality binary silhouettes, extracted from videos recorded in an specialized studio belonging to the audiovisual facilities of the University Jaume I. Ten healthy volunteers, nine males and one female, were required to walk across a green chroma simulating several abnormal gait styles.
Seven different gait styles were simulated, in which movement of limbs and posture of full body were altered to some extent. They were inspired by pathological gaits characteristic of certain neurological diseases such as Parkinson. An eighth style under neutral and unaffected appearance was also included. Each person was recorded twice for each gait pattern, and all sequences were recorded from a lateral view.
Videos were acquired in the Laboratory of Communication Sciences (LABCOM) of the University Jaume I. It is an advanced facility to support educational activities and active research in audiovisual communication technologies, advertising and journalism.
Seven different gait styles were simulated, in which movement of limbs and posture of full body were altered to some extent, inspired by pathological gaits typical of certain neurological diseases. An eigth style, neutral and unaffected, was also provided.
The use of a uniform background, such as a green chroma, facilitated the binarization of frames and extraction of high-quality silhouettes.
A file with full specifications including file and folder names can be downloaded here.