Object Recognition with and without Objects

By Zhuotun Zhu, Lingxi Xie, Alan L. Yuille. 2016

89 citations recorded by the source corpus.

While recent deep neural networks have achieved a promising performance on object recognition, they rely implicitly on the visual contents of the whole image. In this paper, we train deep neural networks on the foreground (object) and background (context) regions of images respectively. Considering human recognition in the same situations, networks trained on the pure background without objects achieves highly reasonable recognition performance that beats humans by a large margin if only given context. However, humans still outperform networks with pure object available, which indicates networks and human beings have different mechanisms in understanding an image. Furthermore, we straightforwardly combine multiple trained networks to explore different visual cues learned by different networks. Experiments show that useful visual hints can be explicitly learned separately and then combined to achieve higher performance, which verifies the advantages of the proposed framework.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Shortcut learning in deep neural networks (2020, background): “If cows happen to be on grassland for most of the training data, detecting grass instead of cows becomes a successful strategy for solving a classification problem in an unintended way; and indeed many models base their predictions on context [23, 24, 25, 26, 9, 27, 10].”
Strike (With) a Pose: Neural Networks Are Easily Fooled by Strange Poses of Familiar Objects (2018, background): “, based on only the background content) [57].”
Distilling Object Detectors via Decoupled Features (2021, background): “Except that, prior literature has shown that there is a strong relationship between objects and background [53, 69].”
Deep convolutional networks do not classify based on global object shape (2018, background): “Typical tests of DCNNs include contextual information, which has been shown to be so important that networks perform reliably better than chance in classifying objects with images in which the object to be classified has been removed [21].”
Deep Nets: What have They Ever Done for Vision? (2018, background): “These methods outperformed the previous best methods, the Deformable Part Models (Felzenszwalb et al., 2010), for the PASCAL object detection challenge (PASCAL was the main object detection and clas-siﬁcation challenge before ImageNet).”
ObjectNet: A large-scale bias-controlled dataset for pushing the limits of object recognition models (2019, background): “[2] train models that separate foregrounds and backgrounds explicitly to become more resilient to biases.”