Heel strike detection based on human walking movement for surveillance analysis

Jung, Sung-Uk; Nixon, Mark S.

doi:10.1016/j.patrec.2012.08.007

Cited by 14 publications

(11 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We successfully detected 254 out of 267 heel strikes in the SOTON and 369 out of 391 in OU-ISIR dataset. Compared with the results of a previous study of detecting heel strikes (95.6% on the SOTON gait database) [5], the detection rate is similar and our approach only requires three consecutive frames. The results in Figure 7(a) also show capability to detect heel strikes in outdoor imagery where the lighting is uncontrolled.…”

Section: Heel Strike Position Verificationsupporting

confidence: 59%

“…According to this clue, we can determine the key frame and position of the heel strike precisely. Previous approaches to heel strike detection have used more of the image sequence and have determined the frame which has the most corners [4] and by detection based on the sinusoidal movement of the head and a silhouette accumulator map [5]. In contrast, the new approach uses only three consecutive frames to detect acceleration and thence heel strikes, and further can be generalised to detect crime events which invariably involve acceleration rather than smooth movement.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Detecting acceleration for gait and crime scene analysis

Sun¹,

Hare²,

Nixon³

2016

7th International Conference on Imaging for Crime Detection and Prevention (ICDP 2016)

View full text Add to dashboard Cite

Identifying criminals from CCTV footage is often a difficult task for crime investigations. The quality of CCTV is often low and criminals can cover their face and wear gloves (to withhold fingerprints) when committing a crime. Gait is the optimal choice in this circumstance since people can be recognised by their walking style, even at a distance with low resolution imagery. The location of the frame when the heel strikes the floor is essential for some gait analyses. We propose a new method to detect heel strikes: by radial acceleration which can also generalise to crime analysis. The frame and position of the heel strikes can be estimated by the quantity and the circle centres of radial acceleration, derived from the optical flow (using DeepFlow). Experimental results show high detection rate on two different gait databases and good robustness under different kinds of noise. We analyse detection of heel strikes to show robustness then we analyse crime scenes to show generalisation capability since violent crime often involves much acceleration. As such, we provide a new basis to a baseline technique in crime scene analysis.

show abstract

Section: Heel Strike Position Verificationsupporting

confidence: 59%

Section: Introductionmentioning

confidence: 99%

Detecting acceleration for gait and crime scene analysis

Sun¹,

Hare²,

Nixon³

2016

7th International Conference on Imaging for Crime Detection and Prevention (ICDP 2016)

View full text Add to dashboard Cite

show abstract

“…The authors used a gait trajectory model (vertical oscillation) to retrieve spatio-temporal information of foot contacts. For walking in a calibrated, biometric tunnel, Jung and Nixon [4] reported that 95.6% of foot contacts were identified within ± 100 mm of ground truth data. However, temporal data were not assessed.…”

Section: Introductionmentioning

confidence: 99%

“…This reflected the sequential accumulation of image data to allow the spatial identification of foot contacts. Jung and Nixon [4] presented a single camera method for the spatial identification of foot contacts during walking. The authors used a gait trajectory model (vertical oscillation) to retrieve spatio-temporal information of foot contacts.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Non-invasive, Spatio-temporal Gait Analysis for Sprint Running Using a Single Camera

Dunn

Kelley

2015

Procedia Engineering

View full text Add to dashboard Cite

Sprint running velocity is the product of step length and step rate. A tool to measure these key metrics would aid sprint training. Athletes require fast and non-invasive analysis tools, to allow them to focus on performance. A non-invasive, single camera gait analysis system (Gait Analyser) was developed and installed at the Sheffield Hallam University City Athletics Stadium (SHUCAS). The Gait Analyser filmed athletes sprinting in lanes 1, 5 and 8 wearing different coloured shoes in varied lighting conditions (e.g. sunlight or overcast). The Gait Analyser automatically identified the position and time of athlete's foot contacts, allowing the calculation of step length, step time and step velocity. Output data were compared to corresponding, manually identified measurements. For optimised setups, 100% of foot contacts were identified. Resultant direction root-mean square error (RMSE) for foot contact position and time was 108.9 mm and 0.03 s respectively. RMSE for step length, step time and step velocity was 4.9 mm, 0.00 s and 0.07 m·s -1 respectively. The Gait Analyser measured spatio-temporal gait parameters of sprint running in situ without applying markers or sensors to the athlete or the running track: results were available 2-3 s after capture.

show abstract

Automatic high fidelity foot contact location and timing for elite sprinting

Evans

Colyer

Salo

et al. 2021

Machine Vision and Applications

View full text Add to dashboard Cite

Making accurate measurements of human body motions using only passive, non-interfering sensors such as video is a difficult task with a wide range of applications throughout biomechanics, health, sports and entertainment. The rise of machine learning-based human pose estimation has allowed for impressive performance gains, but machine learning-based systems require large datasets which might not be practical for niche applications. As such, it may be necessary to adapt systems trained for more general-purpose goals, but this might require a sacrifice in accuracy when compared with systems specifically developed for the application. This paper proposes two approaches to measuring a sprinter’s foot-ground contact locations and timing (step length and step frequency), a task which requires high accuracy. The first approach is a learning-free system based on occupancy maps. The second approach is a multi-camera 3D fusion of a state-of-the-art machine learning-based human pose estimation model. Both systems use the same underlying multi-camera system. The experiments show the learning-free computer vision algorithm to provide foot timing to better than 1 frame at 180 fps, and step length accurate to 7 mm, while the system based on pose estimation achieves timing better than 1.5 frames at 180 fps, and step length estimates accurate to 20 mm.

show abstract

Heel strike detection based on human walking movement for surveillance analysis

Cited by 14 publications

References 14 publications

Detecting acceleration for gait and crime scene analysis

Detecting acceleration for gait and crime scene analysis

Non-invasive, Spatio-temporal Gait Analysis for Sprint Running Using a Single Camera

Automatic high fidelity foot contact location and timing for elite sprinting

Contact Info

Product

Resources

About