Parallelization of the optical flow computation in sequences from moving cameras

García-Dopico, Antonio; Pedraza, José Luís; Nieto, Marcos; Pérez, Aritz; Rodríguez, Santiago; Navas, J.

doi:10.1186/1687-5281-2014-18

Cited by 6 publications

(6 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The estimation of the optical flow in real time is a challenging task because it requires a lot of computation efforts and in the same time the hardware to remain low. There were a lot of works in optimizing optical flow algorithms in CPU [2]- [4], GPU [5]- [7] and FPGAs. Especially for FPGAs, some works use the Lucas-Kanade (L&K) method with mono-scale and multi-scale implementations for [8] while [9] remains on the multi-scale Phase-based algorithm and [10] proposes a lower frame memory access to reduce external memory interactions.…”

Section: A State-of-the-artmentioning

confidence: 99%

FPGA Acceleration of the Horn and Schunck Hierarchical Algorithm

Bournias

Chotin

Lacassagne

2021

2021 IEEE International Symposium on Circuits and Systems (ISCAS)

View full text Add to dashboard Cite

This work proposes a highly tunable motion estimation architecture. We implement the Horn and Schunck algorithm with the hierarchical extension for larger motion estimations in FPGAs. Different architectures are explored dealing with interpolation, pipeline, parallelism and arithmetic format, in order to fit performance. We show in our exploration, how the different cores of our system should be used to increase the throughput. Our smallest design achieves a 30.8 Mpixel/s in a 1024x1024 resolution and the fastest 507 Mpixel/s which is one of the fastest ever achieved, as far as we know, for FPGAs.

show abstract

Section: A State-of-the-artmentioning

confidence: 99%

FPGA Acceleration of the Horn and Schunck Hierarchical Algorithm

Bournias

Chotin

Lacassagne

2021

2021 IEEE International Symposium on Circuits and Systems (ISCAS)

View full text Add to dashboard Cite

show abstract

“…CMLA's IPol website [4] also provides various codes of recent algorithms. Optimized implementations of optical flow algorithms were the subject of numerous works on FPGA [5], [6], [7], [8] and on GPU [9], [10], [11], [12], but few on CPU [11], [13]. It should also be noted that optical flow estimations based on machine learning are gaining in popularity in the scientific community [14], [15].…”

Section: Optical Flow Iterative Algorithmsmentioning

confidence: 99%

Energy and Execution Time Comparison of Optical Flow Algorithms on SIMD and GPU Architectures

Petreto¹,

Hennequin²,

Kœhler³

et al. 2018

2018 Conference on Design and Architectures for Signal and Image Processing (DASIP)

View full text Add to dashboard Cite

This article presents and compares optimized implementations of two optical flow algorithms on several target boards comprising multi-core SIMD processors and GPUs. The two algorithms are Horn-Schunck (HS) and TV-L1, and have been chosen because they are both well-known, and because of their different computational complexity and accuracy. For both algorithms, we have made parallel optimized SIMD implementations, while HS has also been implemented on GPUs. For each algorithm, the comparison between the different versions and target boards is carried out in a two-dimensional fashion: in terms of computing speed-in order to achieve real-time computation-and in terms of energy consumption since we target embedded systems. The results show that for HS, the GPUs are the most efficient in both dimensions, able to process in realtime performances (25 frames per second) up to 8Mpix images for 0.35J per image, against 1.8Mpix images for 0.24J per image on CPU. The results also highlight the impact of optimizations on TV-L1: far slower than HS without optimization, it can almost match its performance after optimization on CPU, and can achieve real-time performances with 0.25J for 1.4Mpix images. We hope these results will help developers design optical flow embedded systems.

show abstract

“…In [32] and [1], we present a parallelization of the Lucas-Kanade algorithm applied to the computation of optical flow on video sequences taken from a moving vehicle in real traffic. These types of images present several sources of optical flow: road objects (lines, trees, houses, panels, etc.…”

Section: Parallelization Of Optical Flowmentioning

confidence: 99%

“…All these processes require highly computing-intensive operations. Due to this fact, a parallel real-time version of this system has been developed and implemented as described in [1]. In this parallel version, all processing stages are carried out at a 45-frames per second (fps) rate when applied to 502 × 288 small images or at a 15-fps rate when highresolution 720 × 576 images are used.…”

Section: Introductionmentioning

confidence: 99%

Locating moving objects in car-driving sequences

et al. 2014

Self Cite

View full text Add to dashboard Cite

This paper presents a system for the search and detection of moving objects in a sequence of images previously captured by a camera installed in a conventional vehicle. The objective is the design and implementation of a software system based on optical flow analysis to detect and identify moving objects as perceived by a driver, taking into account that these objects could interfere with the driver's behavior, either through distraction or by posing an actual danger that may require an active response. The problem presents significant difficulties because the vehicle travels on conventional roads, open to normal traffic. Consequently, the scenes are recorded with natural lighting, i.e., under highly variable conditions (intensity, shadows, etc.). Furthermore, the use of a moving camera makes it difficult to properly identify static objects such as the road itself, signals, buildings, landscapes, and moving objects of the same speed, such as pedestrians or other vehicles. The proposed method consists of three stages. First, the optical flow is calculated for each image of the sequence, as a first estimate of the apparent motion. In a second step, two segmentation processes are addressed: the optical flow itself and the images of the sequence. Finally, in the last stage, the results of these two segmentation processes are combined to obtain the movement of the objects present in the sequence, identifying both their direction and magnitude. The quality of the results obtained with different sequences of real images makes this software suitable for systems to study driver behavior and to help detect danger situations, as various international traffic agencies consider in their research projects.

show abstract

Parallelization of the optical flow computation in sequences from moving cameras

Cited by 6 publications

References 44 publications

FPGA Acceleration of the Horn and Schunck Hierarchical Algorithm

FPGA Acceleration of the Horn and Schunck Hierarchical Algorithm

Energy and Execution Time Comparison of Optical Flow Algorithms on SIMD and GPU Architectures

Locating moving objects in car-driving sequences

Contact Info

Product

Resources

About