Exploring spatial sparsity for more efficient CNNs in semantic segmentation

Add to your list(s) Download to your calendar using vCal

• Giorgos Zampokas
• Monday 19 June 2023, 13:00-14:00
• Department of Electrical and Electronic Engineering, Room 909B.

If you have a question about this talk, please contact George A Constantinides.

CNNs demonstrate state-of-the art performance in most computer vision tasks. However, their improved performance comes with increased computational cost, limiting their adoption and usage in practical applications. To address that, we explore a flexible optimization scheme, which introduces spatial sparsity to a CNN to produce more efficient distribution of calculations. In this seminar, we will discuss how spatial sparsity can be introduced to a CNN and how it affects its performance. Moreover, we will also investigate the sparsity-based optimization scheme and how it can include actual latency to adapt to a target device.

This talk is part of the CAS Talks series.

This talk is included in these lists:

• Andrea Picciau's list
• CAS Talks
• Circuits and Systems Group: Internal Seminars
• Department of Electrical and Electronic Engineering, Room 909B

Note that ex-directory lists are not shown.