Survey on hardware implementations of visual object trackers

El-Shafie, Al-Hussein A.; Habib, S. E. D.

doi:10.1049/iet-ipr.2018.5952

Cited by 15 publications

(9 citation statements)

References 73 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is challenging for a tracker to handle all these variations in a single framework. Therefore, numerous algorithms and schemes exist in literature aiming to tackle the tracking challenges and improve the overall tracing performance [1]- [3]. A typical tracking system consists of two main models, motion model and appearance model.…”

Section: Introductionmentioning

confidence: 99%

Fast CNN-Based Object Tracking Using Localization Layers and Deep Features Interpolation

El-Shafie

Zaki

Habib

2019

2019 15th International Wireless Communications &Amp; Mobile Computing Conference (IWCMC)

View full text Add to dashboard Cite

Object trackers based on Convolution Neural Network (CNN) have achieved state-of-the-art performance on recent tracking benchmarks, while they suffer from slow computational speed. The high computational load arises from the extraction of the feature maps of the candidate and training patches in every video frame. The candidate and training patches are typically placed randomly around the previous target location and the estimated target location respectively. In this paper, we propose novel schemes to speedup the processing of the CNN-based trackers. We input the whole region-of-interest once to the CNN to eliminate the redundant computations of the random candidate patches. In addition to classifying each candidate patch as an object or background, we adapt the CNN to classify the target location inside the object patches as a coarse localization step, and we employ bilinear interpolation for the CNN feature maps as a fine localization step. Moreover, bilinear interpolation is exploited to generate CNN feature maps of the training patches without actually forwarding the training patches through the network which achieves a significant reduction of the required computations. Our tracker does not rely on offline video training. It achieves competitive performance results on the OTB benchmark with 8x speed improvements compared to the equivalent tracker.

show abstract

Section: Introductionmentioning

confidence: 99%

Fast CNN-Based Object Tracking Using Localization Layers and Deep Features Interpolation

El-Shafie

Zaki

Habib

2019

2019 15th International Wireless Communications &Amp; Mobile Computing Conference (IWCMC)

View full text Add to dashboard Cite

show abstract

“…We designed a generic hardware accelerator that performs the eight matrix multiplications. For all the steps, the computation can be viewed as [ ] = [ 1] * [ 2]. The finite state machine reads the appropriate [ 1] and [ 2] in each computation step from the two DRAM3 modules on the evaluation board.…”

Section: B Computation Enginementioning

confidence: 99%

“…Therefore, numerous hardware and hardware/software co-design approaches have been pursued for better tracker implementation on embedded and mobile systems. Reference [1] surveyed the hardware implementation of the object trackers.…”

Section: Introductionmentioning

confidence: 99%

Towards an Efficient Hardware Implementation of CNN-Based Object Trackers

El-Shafie

Zaki

Habib

2021

2021 IEEE International Symposium on Circuits and Systems (ISCAS)

View full text Add to dashboard Cite

Convolution Neural Network (CNN) has achieved a performance boost to the visual tracking field. However, CNN-based trackers feature slow computational speed and large memory size. These issues challenge the embedded implementation of the CNN-based trackers. In this paper, we show how interpolation schemes can significantly reduce the memory requirements. In addition, we present a design-space exploration of the fixed-point representation of the CNN-based trackers aiming for a cost-efficient hardware implementation. Moreover, we present a hardware accelerator of the online training stage of the tracker. Our proposed accelerator can train the whole fully connected layers at a rate of 45.9 frames-per-second.

show abstract

“…The dramatic increase in the computational power has opened the door for the creation of new tracking algorithms to overcome the tracking field challenges [1].…”

mentioning

confidence: 99%

“…OF SYSTEM-LEVEL FEATURES* For all ILNET variants given in this table, the weights and the feature maps are represented by 8-bit for the convolutional layers, and by 16-bit for the fully connected layers 1. No dropout, 2 dropout pattern: 8, dropout pattern: 18, dropout pattern:30 We can see that equation(7) has two exact expoential functions and one logarithmic function.…”

mentioning

confidence: 99%

An Efficient Hardware Implementation of CNN-Based Object Trackers for Real-Time Applications

El-Shafie¹,

Zaki²,

Habib³

2021

Preprint

View full text Add to dashboard Cite

<div>The object tracking research continues to be active since long period because of the several real-world variations imposed in the tracking process, like occlusion, changing appearance, illumination changes and cluttered background. With wide range of applications, embedded implementations are typically pursed for the tracking systems. Although object trackers based on Convolution Neural Network (CNN) have achieved state-of-the-art performance, they challenge the embedded implementations because of slow speed and large memory requirement. In this paper, we address these limitations on the algorithm-side and the circuitside. On the algorithm side, we adopt interpolation schemes which can significantly reduce the processing time and the memory storage requirements. We also evaluate the approximation of the hardware-expensive computations aiming for an efficient hardware implementation. Moreover, we modified the online-training scheme in order to achieve a constant processing time across all video frames. On the circuit side, we developed a hardware accelerator of the online training stage. We avoid the transposed reading from the external memory to speed-up the data movement with no performance degradation. Our proposed hardware accelerator achieves 45.9 frames-per-second in training the fully connected layers.</div>

show abstract

Survey on hardware implementations of visual object trackers

Cited by 15 publications

References 73 publications

Fast CNN-Based Object Tracking Using Localization Layers and Deep Features Interpolation

Fast CNN-Based Object Tracking Using Localization Layers and Deep Features Interpolation

Towards an Efficient Hardware Implementation of CNN-Based Object Trackers

An Efficient Hardware Implementation of CNN-Based Object Trackers for Real-Time Applications

Contact Info

Product

Resources

About