| DC Field | Value | Language |
|---|
| dc.contributor.author | Pirshtuk, D. | - |
| dc.coverage.spatial | Минск | en_US |
| dc.date.accessioned | 2024-02-22T06:24:01Z | - |
| dc.date.available | 2024-02-22T06:24:01Z | - |
| dc.date.issued | 2023 | - |
| dc.identifier.citation | Pirshtuk, D. Low-latency Human Portrait Segmentation Network Optimized for CPU Inference / D. Pirshtuk // Pattern Recognition and Information Processing (PRIP'2023) = Распознавание образов и обработка информации (2023) : Proceedings of the 16th International Conference, October 17–19, 2023, Minsk, Belarus / United Institute of Informatics Problems of the National Academy of Sciences of Belarus. – Minsk, 2023. – P. 186–192. | en_US |
| dc.identifier.uri | https://libeldoc.bsuir.by/handle/123456789/54315 | - |
| dc.description.abstract | In this paper, we discuss a design of fast and lightweight neural networks for working in real-time under
very strict resource constraints and describe a human portrait segmentation method with temporal consistency based on an encoder-decoder architecture with a state-of-the-art CPU optimized PP-LCNet backbone and a custom decoder. Proposed neural network can process about 150-500 frames per second using only a single CPU thread with high accuracy and can be used for virtual background replacements in video conferencing and other augmented reality cases. | en_US |
| dc.language.iso | en | en_US |
| dc.publisher | BSU | en_US |
| dc.subject | материалы конференций | en_US |
| dc.subject | portrait segmentation | en_US |
| dc.subject | image segmentation | en_US |
| dc.subject | neural networks | en_US |
| dc.subject | computer vision | en_US |
| dc.subject | augmented reality | en_US |
| dc.title | Low-latency Human Portrait Segmentation Network Optimized for CPU Inference | en_US |
| dc.type | Article | en_US |
| Appears in Collections: | Pattern Recognition and Information Processing (PRIP'2023) = Распознавание образов и обработка информации (2023)
|
