计算机科学
光学(聚焦)
人工智能
水准点(测量)
卷积(计算机科学)
特征(语言学)
分割
目标检测
卷积神经网络
特征提取
模式识别(心理学)
人工神经网络
对象(语法)
计算机视觉
图像分割
图像(数学)
哲学
物理
光学
地理
语言学
大地测量学
作者
Xizhou Zhu,Han Hu,Stephen Lin,Jifeng Dai
标识
DOI:10.1109/cvpr.2019.00953
摘要
The superior performance of Deformable Convolutional Networks arises from its ability to adapt to the geometric variations of objects. Through an examination of its adaptive behavior, we observe that while the spatial support for its neural features conforms more closely than regular ConvNets to object structure, this support may nevertheless extend well beyond the region of interest, causing features to be influenced by irrelevant image content. To address this problem, we present a reformulation of Deformable ConvNets that improves its ability to focus on pertinent image regions, through increased modeling power and stronger training. The modeling power is enhanced through a more comprehensive integration of deformable convolution within the network, and by introducing a modulation mechanism that expands the scope of deformation modeling. To effectively harness this enriched modeling capability, we guide network training via a proposed feature mimicking scheme that helps the network to learn features that reflect the object focus and classification power of R-CNN features. With the proposed contributions, this new version of Deformable ConvNets yields significant performance gains over the original model and produces leading results on the COCO benchmark for object detection and instance segmentation.
科研通智能强力驱动
Strongly Powered by AbleSci AI