Implicit Feature Pyramid Network for Object Detection

Abstract

In this paper, we present an implicit feature pyramid network (i-FPN) for object detection. Existing FPNs stack several cross-scale blocks to obtain large receptive field. We propose to use an implicit function, recently introduced in deep equilibrium model (DEQ), to model the transformation of FPN. We develop a residual-like iteration to updates the hidden states efficiently. Experimental results on MS COCO dataset show that i-FPN can significantly boost detection performance compared to baseline detectors with ResNet-50-FPN: +3.4, +3.2, +3.5, +4.2, +3.2 mAP on RetinaNet, Faster-RCNN, FCOS, ATSS and AutoAssign, respectively.

Tech report

References in corpus (10)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• YOLOv4: Optimal Speed and Accuracy of Object Detection
• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• Universal Transformers
• FreeAnchor: Learning to Match Anchors for Visual Object Detection
• Deep Equilibrium Models
• AutoAssign: Differentiable Label Assignment for Dense Object Detection
• DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution
• Multiscale Deep Equilibrium Models

Cited by in corpus (3)

• You Only Learn One Representation: Unified Network for Multiple Tasks
• Invisible Gas Detection: An RGB-Thermal Cross Attention Network and A New Benchmark
• On Training Implicit Models