Improving the Performance of the Contextual Spaces Re-Ranking Algorithm on Heterogeneous Systems

• Flávia Pisani UNICAMP
• Daniel Pedronette UNESP
• Ricardo Torres UNICAMP
• Edson Borin UNICAMP

DOI: https://doi.org/10.5753/wscad.2015.14278

Resumo

Re-ranking algorithms have been proposed to improve the effectiveness of Content-Based Image Retrieval (CBIR) systems by exploiting contextual information encoded in distance measures and ranked lists. In this paper, we show how we improved the efficiency of one of these algorithms, called Contextual Spaces Re-Ranking. We propose a modification to the algorithm that reduces its execution time by 1.6x on average and improves its accuracy in most of our test cases. We also parallelized the implementation with OpenCL to use the CPU and GPU of an Accelerated Processing Unit (APU). Employing these devices to run different parts of the code resulted in speedups that range from 3.3x to 4.2x in comparison with the total execution time of the serial version.

Referências

AMD, A. M. D. I. (2013). AMD Accelerated Parallel Processing OpenCLTM Programming Guide. Accessed: July 22, 2015.

Bentley, J. (2000). Programming Pearls. ACM Press Series. Addison-Wesley.

Datta, R., Joshi, D., Li, J., and Wang, J. Z. (2008). Image retrieval: Ideas, inuences, and trends of the new age. ACM Comput. Surv., 40:5:1–5:60.

Ferreira, J. R., Oliveira, M. C., and Freitas, A. L. (2014). Performance Evaluation of Medical Image Similarity Analysis in a Heterogeneous Architecture. In Proc. IEEE CBMS, pages 159–164.

Latecki, L. J., Lakmper, R., and Eckhardt, U. (2000). Shape Descriptors for Non-rigid Shapes with a Single Closed Contour. In Proc. CVPR, pages 424–429.

Pedronette, D. C. G. and da S. Torres, R. (2010). Shape Retrieval using Contour Features and Distance Optmization. In Proc. VISAPP, pages 197–202.

Pedronette, D. C. G. and da S. Torres, R. (2011). Exploiting contextual spaces for image re-ranking and rank aggregation. In Proc. ICMR, pages 13:1–13:8.

Pedronette, D. C. G., da S. Torres, R., Borin, E., and Breternitz, M. (2012). Efcient Image Re-Ranking Computation on GPUs. In Proc. ISPA, pages 95–102.

Pedronette, D. C. G., da S. Torres, R., Borin, E., and Breternitz, M. (2013). Image reranking acceleration on gpus. In Proc. SBAC-PAD, pages 176–183.

Sevilla, J., Bernabe, S., and Plaza, A. (2014). Unmixing-based content retrieval system for remotely sensed hyperspectral imagery on GPUs. J. Supercomput., 70(2):588–599.

Steele, J. and Cochran, R. (2007). Introduction to GPGPU Programming. In Proc. ACMSE 45, pages 508–508.

Stone, J. E., Gohara, D., and Shi, G. (2010). OpenCL: A Parallel Programming Standard for Heterogeneous Computing Systems. Comput. Sci. Eng., 12:66–73.

Teodoro, G., Valle, E., Mariano, N., Torres, R., Meira Jr, W., and Saltz, J. H. (2014). Approximate similarity search for online multimedia services on distributed CPU–GPU platforms. VLDB J., 23(3):427–448.

Wang, J., Li, Y., Bai, X., Zhang, Y., Wang, C., and Tang, N. (2011). Learning contextsensitive similarity by shortest path propagation. Pattern Recognit., 44(10-11):2367– 2374.

Yang, X., Bai, X., Latecki, L. J., and Tu, Z. (2008). Improving Shape Retrieval by Learning Graph Transduction. In Proc. ECCV, volume 4, pages 788–801.

Yang, X. and Latecki, L. J. (2011). Afnity learning on a tensor product graph with applications to shape and image retrieval. In Proc. CVPR, pages 2369–2376.