Iterative Visual Reasoning Beyond Convolutions

被引：154

作者：

Chen, Xinlei ^{[1
]}

Li, Li-Jia ^{[2
]}

Li Fei-Fei ^{[2
]}

Gupta, Abhinav ^{[1
]}

机构：

[1] Carnegie Mellon Univ, Pittsburgh, PA 15213 USA

[2] Google, Mountain View, CA 94043 USA

来源：

2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) | 2018年

关键词：

D O I：

10.1109/CVPR.2018.00756

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We present a novel framework for iterative visual reasoning. Our framework goes beyond current recognition systems that lack the capability to reason beyond stack of convolutions. The framework consists of two core modules: a local module that uses spatial memory [4] to store previous beliefs with parallel updates; and a global graph-reasoning module. Our graph module has three components: a) a knowledge graph where we represent classes as nodes and build edges to encode different types of semantic relationships between them; b) a region graph of the current image where regions in the image are nodes and spatial relationships between these regions are edges; c) an assignment graph that assigns regions to classes. Both the local module and the global module roll-out iteratively and cross-feed predictions to each other to refine estimates. The final predictions are made by combining the best of both modules with an attention mechanism. We show strong performance over plain ConvNets, e.g. achieving an 8.4% absolute improvement on ADE [55] measured by per-class average precision. Analysis also shows that the framework is resilient to missing regions for reasoning.

引用

页码：7239 / 7248

页数：10