RRNet: Repetition-Reduction Network for Energy Efficient Decoder of Depth Estimation
This addresses efficiency challenges in depth estimation for applications on resource-limited devices, representing an incremental improvement over existing lightweight methods.
The paper tackles the problem of resource-constrained depth estimation by introducing RRNet, which uses repetition-reduction blocks to improve efficiency. The result is approximately 3.84 times less energy consumption, 3.06 times less memory usage, and 2.21 times faster speed compared to a baseline network, while outperforming other lightweight architectures.
We introduce Repetition-Reduction network (RRNet) for resource-constrained depth estimation, offering significantly improved efficiency in terms of computation, memory and energy consumption. The proposed method is based on repetition-reduction (RR) blocks. The RR blocks consist of the set of repeated convolutions and the residual connection layer that take place of the pointwise reduction layer with linear connection to the decoder. The RRNet help reduce memory usage and power consumption in the residual connections to the decoder layers. RRNet consumes approximately 3.84 times less energy and 3.06 times less meory and is approaximately 2.21 times faster, without increasing the demand on hardware resource relative to the baseline network (Godard et al, CVPR'17), outperforming current state-of-the-art lightweight architectures such as SqueezeNet, ShuffleNet, MobileNet and PyDNet.