SRF-GAN: Super-Resolved Feature GAN for Multi-Scale Representation
This work addresses a specific bottleneck in convolutional object detectors for computer vision applications, offering an incremental improvement by replacing interpolation with a GAN-based method.
The paper tackles the problem of noisy and blurred features in multi-scale object detection caused by simple interpolation methods during top-down feature propagation, proposing SRF-GAN to super-resolve features and improving detection accuracy over several recent detectors.
Recent convolutional object detectors exploit multi-scale feature representations added with top-down pathway in order to detect objects at different scales and learn stronger semantic feature responses. In general, during the top-down feature propagation, the coarser feature maps are upsampled to be combined with the features forwarded from bottom-up pathway, and the combined stronger semantic features are inputs of detector's headers. However, simple interpolation methods (e.g. nearest neighbor and bilinear) are still used for increasing feature resolutions although they cause noisy and blurred features. In this paper, we propose a novel generator for super-resolving features of the convolutional object detectors. To achieve this, we first design super-resolved feature GAN (SRF-GAN) consisting of a detection-based generator and a feature patch discriminator. In addition, we present SRF-GAN losses for generating the high quality of super-resolved features and improving detection accuracy together. Our SRF generator can substitute for the traditional interpolation methods, and easily fine-tuned combined with other conventional detectors. To prove this, we have implemented our SRF-GAN by using the several recent one-stage and two-stage detectors, and improved detection accuracy over those detectors. Code is available at https://github.com/SHLee-cv/SRF-GAN.