Split Computing for Complex Object Detectors: Challenges and Preliminary Results
This addresses the problem of efficient inference for object detection in mobile and edge computing scenarios, but it is incremental as it builds on prior split computing work by extending it to a new task.
The paper tackles the challenge of applying split computing to complex object detectors like R-CNN on large datasets such as COCO 2017, finding that naive methods do not reduce inference time, but it demonstrates potential by injecting small bottlenecks.
Following the trends of mobile and edge computing for DNN models, an intermediate option, split computing, has been attracting attentions from the research community. Previous studies empirically showed that while mobile and edge computing often would be the best options in terms of total inference time, there are some scenarios where split computing methods can achieve shorter inference time. All the proposed split computing approaches, however, focus on image classification tasks, and most are assessed with small datasets that are far from the practical scenarios. In this paper, we discuss the challenges in developing split computing methods for powerful R-CNN object detectors trained on a large dataset, COCO 2017. We extensively analyze the object detectors in terms of layer-wise tensor size and model size, and show that naive split computing methods would not reduce inference time. To the best of our knowledge, this is the first study to inject small bottlenecks to such object detectors and unveil the potential of a split computing approach. The source code and trained models' weights used in this study are available at https://github.com/yoshitomo-matsubara/hnd-ghnd-object-detectors .