用地理感知的条件计算，让设备端动物识别更高效。
Efficient Mixture of Geographical Species for On Device Wildlife Monitoring
- 根据地理位置动态启用不同子网络，实现轻量化推理
- 在iNaturalist和iWildcam上实现比基线更低的计算开销
- 适合边缘设备部署的野生动物监测场景
高效的设备端模型对近传感器洞察生成具有吸引力，尤其适用于生态保护领域。尽管深度学习研究者正致力于开发低算力模型，但视觉变换器在边缘场景仍属新兴，尤其是基于输入数据的条件执行子网络尚未充分探索。本文提出一种单物种检测器，通过地理感知的条件计算机制，引导结构化子网络的选择。我们提出了按地理位置剪枝专家模型的方法，并在两个地理分布的数据集iNaturalist和iWildcam上验证了条件计算的有效性。
原文摘要 · Abstract (English)
Efficient on-device models have become attractive for near-sensor insight generation, of particular interest to the ecological conservation community. For this reason, deep learning researchers are proposing more approaches to develop lower compute models. However, since vision transformers are very new to the edge use case, there are still unexplored approaches, most notably conditional execution of subnetworks based on input data. In this work, we explore the training of a single species detector which uses conditional computation to bias structured sub networks in a geographically-aware manner. We propose a method for pruning the expert model per location and demonstrate conditional computation performance on two geographically distributed datasets: iNaturalist and iWildcam.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。