A high throughput fully parallel-pipelined FPGA accelerator for dense cloud motion analysis

Johnson, Bibin; Rani, J. Sheeba

doi:10.1109/tencon.2016.7848505

Cited by 3 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the work [34] an FPGA platform was used as an accelerator for cloud motion analysis using the Horn-Schunck algorithm. According to the authors, the proposed application is theoretically capable of processing video streams of up to 3750 × 3750 pixels in real time.…”

Section: Horn-schunck Fpga Implementationsmentioning

confidence: 99%

“…In summary, just like in the case of the LK algorithm, there are multiple solutions of the HS algorithm that work in real time on FPGA platforms, the most important parameters of which are gathered in Table 2. Works such as [33,34,36] were realised for ultra high-resolution video streams, enabling the detection of smaller objects, but they lack in the detection of bigger pixel displacements because of the used single-scale approach. Moreover, in these implementations, the number of iterations of the HS algorithm is low due to the high use of resources in every iteration.…”

Section: Horn-schunck Fpga Implementationsmentioning

confidence: 99%

See 1 more Smart Citation

Real-Time Efficient FPGA Implementation of the Multi-Scale Lucas-Kanade and Horn-Schunck Optical Flow Algorithms for a 4K Video Stream

Błachut

Kryjak

2022

Sensors

View full text Add to dashboard Cite

The information about optical flow, i.e., the movement of pixels between two consecutive images from a video sequence, is used in many vision systems, both classical and those based on deep neural networks. In some robotic applications, e.g., in autonomous vehicles, it is necessary to calculate the flow in real time. This represents a challenging task, especially for high-resolution video streams. In this work, two gradient-based algorithms—Lucas–Kanade and Horn–Schunck—were implemented on a ZCU 104 platform with Xilinx Zynq UltraScale+ MPSoC FPGA. A vector data format was used to enable flow calculation for a 4K (Ultra HD, 3840 × 2160 pixels) video stream at 60 fps. In order to detect larger pixel displacements, a multi-scale approach was used in both algorithms. Depending on the scale, the calculations were performed for different data formats, allowing for more efficient processing by reducing resource utilisation. The presented solution allows real-time optical flow determination in multiple scales for a 4K resolution with estimated energy consumption below 6 W. The algorithms realised in this work can be a component of a larger vision system in advanced surveillance systems or autonomous vehicles.

show abstract

Section: Horn-schunck Fpga Implementationsmentioning

confidence: 99%

Section: Horn-schunck Fpga Implementationsmentioning

confidence: 99%