Multi-scale Recognition with DAG-CNNs
Multi-scale Recognition with DAG-CNNs
复制标题
DOI：
10.1109/iccv.2015.144
复制
发表时间：
2015-05
期刊：
影响因子：
--
通讯作者：
Songfan Yang;Deva Ramanan
中科院分区：
文献类型：
--
作者：
Songfan Yang;Deva Ramanan
我们探索多尺度卷积神经网络（cnn）用于图像分类。当代的方法是从单个输出层提取特征。通过从多层提取特征，可以在分类过程中同时推断出高、中、低层特征。由此产生的多尺度体系结构本身可以被视为一个前馈模型，该模型被结构为一个有向无环图（dag - cnn）。我们使用dag - cnn来学习一组可以在粗粒度和细粒度分类任务之间有效共享的多尺度特征。虽然微调这样的模型有助于性能，但我们表明，即使是“off- self”的多尺度特征也表现得相当好。我们在三个标准场景基准（SUN397， MIT67和Scene15）上进行了广泛的分析和演示了最先进的分类性能。在经过严格基准测试的MIT67和Scene15数据集上，我们的结果将先前报告的最低误差分别降低了23.9%和9.5%。
We explore multi-scale convolutional neural nets (CNNs) for image classification. Contemporary approaches extract features from a single output layer. By extracting features from multiple layers, one can simultaneously reason about high, mid, and low-level features during classification. The resulting multi-scale architecture can itself be seen as a feed-forward model that is structured as a directed acyclic graph (DAG-CNNs). We use DAG-CNNs to learn a set of multi-scale features that can be effectively shared between coarse and fine-grained classification tasks. While fine-tuning such models helps performance, we show that even "off-the-self" multi-scale features perform quite well. We present extensive analysis and demonstrate state-of-the-art classification performance on three standard scene benchmarks (SUN397, MIT67, and Scene15). In terms of the heavily benchmarked MIT67 and Scene15 datasets, our results reduce the lowest previously-reported error by 23.9% and 9.5%, respectively.