Edge Device Implementation - Artificial Intelligence (AI) Development
Design an optimal processing architecture for the target hardware from the trained algorithms, enabling fast execution on edge devices.
By simply providing the algorithm developed by the customer, we will interpret its content and devise an architecture suitable for the algorithm. We support various frameworks such as Caffe, Keras, TensorFlow, and PyTorch. Drawing on decades of experience in digital circuit design and embedded software development, we will realize optimal solutions without burdening the customer. To implement algorithms on devices such as FPGAs and GPUs in edge environments, it is necessary to have a thorough understanding of the target device's characteristics and resources, which may require architectural design and, in some cases, network optimization. We leverage our implementation experience on devices including FPGAs, GPUs (Jetson AGX Xavier), and AI chips such as TPUs (Google TPU) and VPUs (Intel Movidius Myriad X VPU) to meet the customer's needs from device selection to implementation. <Achievements> ○ Pose estimation ○ General object detection ○ In-vehicle area segmentation
- Company：三栄ハイテックス
- Price：Other
