Deep Neural Network for Structural Prediction and Lane Detection in Traffic Scene

Abstract

Hierarchical neural networks have been shown to be effective in learning representative image features and recognizing object classes. However, most existing networks combine the low/middle level cues for classification without accounting for any spatial structures. For applications such as understanding a scene, how the visual cues are spatially distributed in an image becomes essential for successful analysis. This paper extends the framework of deep neural networks by accounting for the structural cues in the visual signals. In particular, two kinds of neural networks have been proposed. First, we develop a multitask deep convolutional network, which simultaneously detects the presence of the target and the geometric attributes (location and orientation) of the target with respect to the region of interest. Second, a recurrent neuron layer is adopted for structured visual detection. The recurrent neurons can deal with the spatial distribution of visible cues belonging to an object whose shape or structure is difficult to explicitly define. Both the networks are demonstrated by the practical task of detecting lane boundaries in traffic scenes. The multitask convolutional neural network provides auxiliary geometric information to help the subsequent modeling of the given lane structures. The recurrent neural network automatically detects lane boundaries, including those areas containing no marks, without any explicit prior knowledge or secondary modeling.

Learning Multilayer Channel Features for Pedestrian Detection

“…Recently, deep Convolutional Neural Network (CNN) based methods have also achieved great success in object detection [26], [15], [19], [29], [18], [27], [28]. Generally speaking, it firstly generates the candidate object proposals [17], [16] and then uses the trained CNN model [26], [15] to classify these proposals.…”

Multi-lane Detection Using Instance Segmentation and Attentive Voting

“…A variety of lane detection methods have been proposed starting from the traditional methods using handcrafted features ( [1], [2], [3], [4], [5], [6]) to the modern state of the art end-to-end trainable deep architectures ( [7], [8], [9], [10], [11], [12]). Spatial CNN [11] and endto-end lane segmentation [12] use CNN-based approach to exploit the spatial information in a road scene and try to develop an understanding of a road scene.…”

Recognition of Damaged Arrow-Road Markings by Visible Light Camera Sensor Based on Convolutional Neural Network

“…Convolution can allow image-recognition networks to function in a manner similar to biological systems and produce more accurate results [20]. In recent works, a CNN has also been used for detection and classification of traffic signs [21], lane detection [22], and lane position estimation [23]. However, there is no previous research documenting studies of arrow-road marking recognition based on a CNN.…”