Fast GraspNeXt：一种用于边缘机器人抓取计算机视觉任务多任务学习的快速自注意力神经网络架构
Fast GraspNeXt: A Fast Self-Attention Neural Network Architecture for Multi-task Learning in Computer Vision Tasks for Robotic Grasping on the Edge
AI总结：
本文提出 Fast GraspNeXt 自注意力架构，通过带约束的生成式架构搜索平衡多任务性能与边缘推理效率，在 MetaGraspNet 上以更少参数和计算量取得最优结果并显著加速。
摘要：
多任务学习在提升用于机器人抓取的深度学习视觉系统性能方面展现出巨大潜力。然而，较高的架构与计算复杂度会导致其不适合部署在嵌入式设备上，而现实制造和仓库环境中的机械臂通常使用这类设备。因此，为在制造环境中广泛应用，迫切需要设计面向边缘机器人抓取计算机视觉任务的高效多任务深度神经网络架构。受此启发，我们提出 Fast GraspNeXt，这是一种快速自注意力神经网络架构，专为机器人抓取计算机视觉任务中的嵌入式多任务学习而定制。为构建 Fast GraspNeXt，我们采用生成式网络架构搜索策略，并结合一组定制的架构约束，以在多任务学习性能与嵌入式推理效率之间实现强有力的平衡。在 MetaGraspNet 基准数据集上的实验结果表明，与其他高效多任务网络架构设计相比，Fast GraspNeXt 网络设计在多个计算机视觉任务上取得最高性能（平均精度（AP）、准确率和均方误差（MSE）），同时仅具有 17.8M 参数（约小 >5 倍）、259 GFLOPs（低至 >5 倍以下），并且在 NVIDIA Jetson TX2 嵌入式处理器上速度最高快 >3.15 倍。
英文摘要：
Multi-task learning has shown considerable promise for improving the performance of deep learning-driven vision systems for the purpose of robotic grasping. However, high architectural and computational complexity can result in poor suitability for deployment on embedded devices that are typically leveraged in robotic arms for real-world manufacturing and warehouse environments. As such, the design of highly efficient multi-task deep neural network architectures tailored for computer vision tasks for robotic grasping on the edge is highly desired for widespread adoption in manufacturing environments. Motivated by this, we propose Fast GraspNeXt, a fast self-attention neural network architecture tailored for embedded multi-task learning in computer vision tasks for robotic grasping. To build Fast GraspNeXt, we leverage a generative network architecture search strategy with a set of architectural constraints customized to achieve a strong balance between multi-task learning performance and embedded inference efficiency. Experimental results on the MetaGraspNet benchmark dataset show that the Fast GraspNeXt network design achieves the highest performance (average precision (AP), accuracy, and mean squared error (MSE)) across multiple computer vision tasks when compared to other efficient multi-task network architecture designs, while having only 17.8M parameters (about >5x smaller), 259 GFLOPs (as much as >5x lower) and as much as >3.15x faster on a NVIDIA Jetson TX2 embedded processor.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
