Modular Quantization-Aware Training for 6D Object Pose Estimation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913570890973184 |
|---|---|
| author | Javed, Saqib Li, Chengkun Price, Andrew Hu, Yinlin Salzmann, Mathieu |
| author_facet | Javed, Saqib Li, Chengkun Price, Andrew Hu, Yinlin Salzmann, Mathieu |
| contents | Edge applications, such as collaborative robotics and spacecraft rendezvous, demand efficient 6D object pose estimation on resource-constrained embedded platforms. Existing 6D pose estimation networks are often too large for such deployments, necessitating compression while maintaining reliable performance. To address this challenge, we introduce Modular Quantization-Aware Training (MQAT), an adaptive and mixed-precision quantization-aware training strategy that exploits the modular structure of modern 6D pose estimation architectures. MQAT guides a systematic gradated modular quantization sequence and determines module-specific bit precisions, leading to quantized models that outperform those produced by state-of-the-art uniform and mixed-precision quantization techniques. Our experiments showcase the generality of MQAT across datasets, architectures, and quantization algorithms. Remarkably, MQAT-trained quantized models achieve a significant accuracy boost (>7%) over the baseline full-precision network while reducing model size by a factor of 4x or more. Our project website is at: https://saqibjaved1.github.io/MQAT_/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2303_06753 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Modular Quantization-Aware Training for 6D Object Pose Estimation Javed, Saqib Li, Chengkun Price, Andrew Hu, Yinlin Salzmann, Mathieu Computer Vision and Pattern Recognition Machine Learning Robotics Edge applications, such as collaborative robotics and spacecraft rendezvous, demand efficient 6D object pose estimation on resource-constrained embedded platforms. Existing 6D pose estimation networks are often too large for such deployments, necessitating compression while maintaining reliable performance. To address this challenge, we introduce Modular Quantization-Aware Training (MQAT), an adaptive and mixed-precision quantization-aware training strategy that exploits the modular structure of modern 6D pose estimation architectures. MQAT guides a systematic gradated modular quantization sequence and determines module-specific bit precisions, leading to quantized models that outperform those produced by state-of-the-art uniform and mixed-precision quantization techniques. Our experiments showcase the generality of MQAT across datasets, architectures, and quantization algorithms. Remarkably, MQAT-trained quantized models achieve a significant accuracy boost (>7%) over the baseline full-precision network while reducing model size by a factor of 4x or more. Our project website is at: https://saqibjaved1.github.io/MQAT_/ |
| title | Modular Quantization-Aware Training for 6D Object Pose Estimation |
| topic | Computer Vision and Pattern Recognition Machine Learning Robotics |
| url | https://arxiv.org/abs/2303.06753 |