Presentation is loading. Please wait.

Presentation is loading. Please wait.

Adam Wagner Kevin Forbes. Motivation  Take advantage of GPU architecture for highly parallel data-intensive application  Enhance image segmentation.

Similar presentations


Presentation on theme: "Adam Wagner Kevin Forbes. Motivation  Take advantage of GPU architecture for highly parallel data-intensive application  Enhance image segmentation."— Presentation transcript:

1 Adam Wagner Kevin Forbes

2 Motivation  Take advantage of GPU architecture for highly parallel data-intensive application  Enhance image segmentation using Microsoft Kinect IR depth images  Reduce frame-to-frame segmentation overhead with optical flow and iterative simulated annealing  “Depth-supported real-time video segmentation with the Kinect” Algorithm uses Potts model and Metropolis method for segmentation on GPU

3 Implementation No base source code Software frameworks:  OpenCV – image capture, transformations, optical flow  OpenNI – Kinect middleware  CUDA – NVIDIA GPGPU driven architecture Testbed: rcl1.engr.arizona.edu  CPU: Quad-core Intel Xeon 5160, 3.0GHz  GPU: NVIDIA GeForce GTX 480 480 CUDA Cores GDDR5 Threads/block = 1024 Shared memory / block = 48KB

4 Methodology  Primary effort focused on parallelization of segmentation algorithm Without source, code was written from scratch for CPU, then parallelized Memory indexing rearranged to improve coalescing of global loads/stores Much later in semester, some code became available from paper authors  Image divided into thread blocks on GPU Image data loaded into block shared memory from global memory Each thread performs state update on a single pixel

5 Results Input Image: 512 x 384 RGB to HSV Conversion 2000 Metropolis Iterations DimensionsSizeCPU rate (ms)GPU rate (ms) 128 x 961228810.50.054 256 x 19249152410.157 512 x 3841966081690.54 1024 x 7687864328602.115

6 Conclusions  Parallelized algorithm shows vast improvement over CPU version Makes real-time video processing a possibility  Implementation does not match paper More improvement possible through use of simpler data types Still more fine tuned memory arrangement Increase work done by each thread

7


Download ppt "Adam Wagner Kevin Forbes. Motivation  Take advantage of GPU architecture for highly parallel data-intensive application  Enhance image segmentation."

Similar presentations


Ads by Google