Containerization on a self-supervised active foveated approach to computer vision

  • Dario Dematties Northwestern Argonne Institute of Science and Engineering
  • Silvio Rizzi Argonne National Laboratory
  • George K. Thiruvathukal Loyola University Chicago
Palabras clave: Singularity Containerization, NVIDIA DALI, Data Loading and Pre-processing Library, High Performance Computing, Strong Scaling

Resumen

Scaling complexity and appropriate data sets availability for training current Computer Vision (CV) applications poses major challenges. We tackle these challenges finding inspiration in biology and introducing a Self-supervised (SS) active foveated approach for CV. In this paper we present our solution to achieve portability and reproducibility by means of containerization utilizing Singularity. We also show the parallelization scheme used to run our models on ThetaGPU–an Argonne Leadership Computing Facility (ALCF) machine of 24 NVIDIA DGX A100 nodes. We describe how to use mpi4py to provide DistributedDataParallel (DDP) with all the needed information about world size as well as global and local ranks. We also show our dual pipe implementation of a foveator using NVIDIA Data Loading Library (DALI). Finally we conduct a series of strong scaling tests on up to 16 ThetaGPU nodes (128 GPUs), and show some variability trends in parallel scaling efficiency.

Referencias bibliográficas

Alahyane, N., Lemoine- Lardennois, C., Tailhefer, C., Collins, T., Fagard, J., & Doré-Mazars, K. (2016, January). Development and learning of saccadic eye movements in 7- to 42-month-old children. Journal of Vision, 16(1), 6, 1-12. https://doi.org/10.1167/16.1.6

Canfield, R. L., & Haith, M. M. (1991). Young infants' visual expectations for symmetric and asymmetric stimulus sequences. Developmental Psychology, 27(2), 198-208. https://doi.org/10.1037/0012-1649.27.2.198

Canfield, R. L., & Kirkham, N. Z. (2001). Infant Cortical Development and the Prospective Control of Saccadic Eye Movements. Infancy, 2(2), 197-211. https://doi.org/10.1207/S15327078IN0202_5

Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., & Zagoruyko, S. (2020, May 28). arXiv:2005.12872v3 [cs.CV]. End-to-End Object Detection with Transformers. https://doi.org/10.48550/arXiv.2005.12872

Castro, D. C., Walker, I., & Glocker, B. (2020). Causality matters in medical imaging. Nature Communications, 11(3673), 1-10. https://doi.org/10.1038/s41467-020-17478-w

Castro, M., Expósito-Casas, E., López-Martín, E., Lizasoain, L., Navarro-Asencio, E., & Gaviria, J. L. (2015, February). Parental involvement on student academic achievement: A meta-analysis. Educational Research Review, 14, 33-46. https://doi.org/10.1016/j.edurev.2015.01.002

Chen, T., Kornblith, S., Norouzi, M., & Hinton, G. (2020, February 13). A Simple Framework for Contrastive Learning of Visual Representations. In H. Daumé III, & A. Singh (Ed.), Proceedings of the 37 th International Conference on Machine Learning, PMLR 119, 119, pp. 1597-1607. Vienna, Austria. https://doi.org/10.48550/arXiv.2002.05709

Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., . . . Houlsby, N. (2021, October 22). An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. International Conference on Learning Representations ICLR 2021 (pp. 1-21). Vienna, Austria: OpenReview. https://doi.org/10.48550/arXiv.2010.11929

Hikosaka, O., Nakamura, K., & Nakahara, H. (2006). Basal Ganglia Orient Eyes to Reward. Journal of Neurophysiology, 95(2), 567-584. https://doi.org/10.1152/jn.00458.2005

Hsiao, J. H.-W., & Cottrell, G. (2008). Two Fixations Suffice in Face Recognition. Psychological Science, 19(10), 998-1006. https://www.jstor.org/stable/40064836

Ikeda, T., & Hikosaka, O. (2003, August 14). Reward-Dependent Gain and Bias of Visual Responses in Primate Superior Colliculus. Neuron, 39(4), 693-700. https://doi.org/10.1016/S0896-6273(03)00464-1

Johnson, M. H. (1995). The inhibition of automatic saccades in early infancy. Developmental Psychobiology, 28(5), 281-291. https://doi.org/10.1002/dev.420280504

Kato, M., Miyashita, N., Hikosaka, O., Matsumura, M., Usui, S., & Kori, A. (1995, January). Eye Movements in Monkeys with Local Dopamine Depletion in the Caudate Nucleus. I. Deficits in Spontaneous Saccades. The Journal of Neuroscience, 15(1), 912-927. https://doi.org/10.1523/JNEUROSCI.15-01-00912.1995

Preuss, M. (2018, December). Updated: 2018-12-27T08:37:12+00:00, Editorial: What is Edge Computing: The Network Edge Explained. (J. Leavitt, Ed.) Cloudswards Web site: https://www.cloudwards.net/what-is-edge-computing/

Provis, J. M., Diaz, C. M., & Dreher, B. (1998, March). Ontogeny of the primate fovea:a central issue in retinal development. Progress in Neurobiology, 54(5), 549-581. https://doi.org/10.1016/S0301-0082(97)00079-8

Purves, D., Augustine, G. J., Fitzpatrick, D., Hall, W. C., Lamantia, A.-S., McNamara, J. O., & Williams, S. M. (Eds.). (2004). Neuroscience (Third ed.). Sunderland, Massachusetts, USA: Sinauer Associates. https://pages.ucsd.edu/~mboyle/COGS107a/pdf-files/Neuroscience.pdf

Ross-Sheehy, S., Reynolds, E., & Eschman, B. (2020). Evidence for Attentional Phenotypes in Infancy and Their Role in Visual Cognitive Performance. Brain Science, 10(9), 605, 1-24. https://doi.org/10.3390/brainsci10090605

Ross-Sheehy, S., Schneegans, S., & Spencer, J. P. (2015). The Infant Orienting With Attention Task: Assessing the Neural Basis of Spatial Attention in Infancy. Infancy, 20(5), 467-506. https://doi.org/10.1111/infa.12087

Salapatek, P., Aslin, R. N., Simonson, J., & Pulos, E. (1980, December). Infant Saccadic Eye Movements to Visible and Previously Visible Targets. Child Development, 51(4), 1090-1094. https://doi.org/10.2307/1129548

Spotorno, S., Malcolm, G. L., & Tatler, B. W. (2014, February). How context information and target information guide the eyes from the first epoch of search in real-world scenes. Journal of Vision, 14(2), 7, 1-21. https://doi.org/10.1167/14.2.7

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., . . . Polosukhin, I. (2017, June 12). Attention Is All You Need. arXiv(1706.03762 [cs.CL]), 15. https://doi.org/10.48550/arXiv.1706.03762

Weber, R. B., & Daroff, R. B. (1972, March). Corrective movements following refixation saccades: Type and control system analysis. Vision Research, 12(3), 467-475. https://doi.org/10.1016/0042-6989(72)90090-9

Cómo citar
Dematties, D., Rizzi, S., & Thiruvathukal, G. K. (2024). Containerization on a self-supervised active foveated approach to computer vision. Revista Colombiana De Computación, 25(1), 29–38. Recuperado a partir de https://revistas.unab.edu.co/index.php/rcc/article/view/5055

Descargas

Los datos de descargas todavía no están disponibles.
Publicado
2024-06-30
Sección
Artículo de investigación científica y tecnológica

Métricas

QR Code
Crossref Cited-by logo