Abstract:Region proposal-based detectors, such as Region-Convolutional Neural Networks (R-CNNs), Fast R-CNNs, Faster R-CNNs, and Region-Based Fully Convolutional Networks (R-FCNs), employ a two-stage process involving region proposal generation followed by classification. This approach is effective but computationally intensive and typically slower than proposal-free methods. Therefore, region proposal-free detectors are becoming popular to balance accuracy and speed. This paper proposes a proposal-free, fully convolutional network (PF-FCN) that outperforms other state-of-the-art, proposal-free methods. Unlike traditional region proposal-free methods, PF-FCN can generate a "box map" based on regression training techniques. This box map comprises a set of vectors, each designed to produce bounding boxes corresponding to the positions of objects in the input image. The channel and spatial contextualized sub-network are further designed to learn a "box map". In comparison to renowned proposal-free detectors such as CornerNet, CenterNet, and You Look Only Once (YOLO), PF-FCN utilizes a fully convolutional, single-pass method. By reducing the need for fully connected layers and filtering center points, the method considerably reduces the number of trained parameters and optimizes the scalability across varying input sizes. Evaluations of benchmark datasets suggest the effectiveness of PF-FCN: the proposed model achieved an mAP of 89.6% on PASCAL VOC 2012 and 71.7% on MS COCO, which are higher than those of the baseline Fully Convolutional One-Stage Detector (FCOS) and other classical proposal-free detectors. The results prove the significance of proposal-free detectors in both practical applications and future research.

Geospatial Object Detection Via Deconvolutional Region Proposal Network

DeRPN: Taking a Further Step Toward More General Object Detection.

Deconv R-Cnn For Small Object Detection On Remote Sensing Images

Geospatial Object Detection with Single Shot Anchor-Free Network

Rich Features and Precise Localization with Region Proposal Network for Object Detection.

RFP-Net: Receptive field-based proposal generation network for object detection

Deep Adaptive Proposal Network for Object Detection in Optical Remote Sensing Images

An Enhanced Region Proposal Network for Object Detection Using Deep Learning Method.

C-FCN: Corners-based Fully Convolutional Network for Visual Object Detection

Airport Detection Method with Improved Region-Based Convolutional Neural Network

Deep Adaptive Proposal Network in Optical Remote Sensing Images Objective Detection

Enhancing representation learning by exploiting effective receptive fields for object detection

Oriented Object Detection in Remote Sensing Images with Anchor-Free Oriented Region Proposal Network

Multi-modal 3D object detection by 2D-guided precision anchor proposal and multi-layer fusion

Proposal-Free Fully Convolutional Network: Object Detection Based on a Box Map

Reinforcedet - Object Detection by Integrating Reinforcement Learning with Decoupled Pipeline.

Revisiting Faster R-Cnn: A Deeper Look At Region Proposal Network

Domain-Invariant Region Proposal Network for Cross-Domain Detection

DRPDDet: Dynamic Rotated Proposals Decoder for Oriented Object Detection

A new two-stage object detection network without RoI-Pooling

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks