Performance Comparison of KLT and CAMSHIFT Algorithms for Video Object Tracking

Sharma, Prateek; Kokare, Pranjali M.; Kolekar, Maheshkumar H.

doi:10.1007/978-981-13-2685-1_31

Cited by 5 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Also, detecting the bare hand in each video frame is not a rational way to form the intended gesture trajectory. Researchers (Dixit & Agrawal, 2015; Henriques et al, 2015; Sharma et al, 2019) have started exploring tracking approaches such as histogram‐based tracking, Kalman filter, mean‐shift, Kanade‐Lucas‐Tomasi (KLT) algorithm, and so on, for object tracking. However, these models comprise their deficiencies in the pose, position, illumination, speed, scale and occlusion variation.…”

Section: Introductionmentioning

confidence: 99%

“…Complexity in terms of interclass and intraclass variations increases due to natural gesticulation. These are some reasons for researchers to make the gesticulation be performed in a controlled manner, with the pre‐defined pattern, size, style, and scale, using either upper or lowercase characters (Dixit & Agrawal, 2015; Ronneberger et al, 2015; Sharma et al, 2019). These all the challenges need to be explored.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Design and development of a vision‐based system for detection, tracking and recognition of isolated dynamic bare hand gesticulated characters

et al. 2022

View full text Add to dashboard Cite

Detection and tracking are the vital stages to form the gesture trajectory in gesture recognition. It becomes more challenging when the variations in illumination, pose, position, occlusion, scale, speed, blurring effect and complex environment are introduced. Additionally, the background feature domination effect affects the existing deep learning models. A semantic segmentation model is implemented in this work to detect the bare hand to overcome these challenges. A pre‐trained network VGG‐16 is utilized by training with the proposed NITS S‐Net database. Evaluation of the SegNet model is done on EgoHands, Oxford and OUHands databases. To track the bare hand, a SegNet‐based detection and tracking approach is proposed using Kalman filter and point‐tracker. This model achieves 97.01% accuracy (a relative improvement of ~8% from the baseline models) at 0.068 s per frame computational time on NITS hand gesture database VIIIB. The gesticulated characters, that is, alphabets, numbers, operators, special characters, are gesticulated without any constraints on the pattern/strokes. To recognize these 95 multi‐stroke gestures, a deep convolutional neural network (DCNN) is presented using AlexNet. The DCNN model achieves 97.60% (a relative improvement of ~14% from the baseline models) accuracy on the NITS hand gesture database VIIIB merged. Evaluation of the handwritten EMNIST merged (balanced) database resulted in average recognition accuracy of 91.60%.

show abstract

Section: Introductionmentioning

confidence: 99%