Real-Time Deep ConvNet-Based Vehicle Detection Using 3D-LIDAR Reflection Intensity Data

Asvadi, Alireza; Garrote, Luís; Premebida, Cristiano; Peixoto, Paulo; Nunes, Urbano

doi:10.1007/978-3-319-70836-2_39

Cited by 18 publications

(15 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we describe previous methods related to research on conventional interpolation of LiDAR 3D reflection intensity [8,12,13] and deep-learning-based color image generation [18,19,20,21,22,23,24,25,26,27].…”

Section: Related Workmentioning

confidence: 99%

“…Chen et al [12] applied the 2D interpolated reflection intensity image with a camera-based RGB image for lane detection. Asvadi et al [13] also applied the interpolated reflection-intensity image with a color image for vehicle detection and compared the natural neighbor, nearest neighbor, and bilinear interpolation methods. The nearest neighbor interpolation method has the best performance for vehicle detection.…”

Section: Related Workmentioning

confidence: 99%

“…In the same way, the methods produce poor performance if there are many shadows, even during the day. Another approach for solving sparsity is directly to apply the interpolation methods to 2D reflection intensity data [8,12,13]. Because the interpolated image is very noisy, it does not have enough quality for use with conventional computer vision algorithms for object detection and recognition.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Asymmetric Encoder-Decoder Structured FCN Based LiDAR to Color Image Generation

Kim

Yoo

Park

et al. 2019

Sensors

View full text Add to dashboard Cite

In this paper, we propose a method of generating a color image from light detection and ranging (LiDAR) 3D reflection intensity. The proposed method is composed of two steps: projection of LiDAR 3D reflection intensity into 2D intensity, and color image generation from the projected intensity by using a fully convolutional network (FCN). The color image should be generated from a very sparse projected intensity image. For this reason, the FCN is designed to have an asymmetric network structure, i.e., the layer depth of the decoder in the FCN is deeper than that of the encoder. The well-known KITTI dataset for various scenarios is used for the proposed FCN training and performance evaluation. Performance of the asymmetric network structures are empirically analyzed for various depth combinations for the encoder and decoder. Through simulations, it is shown that the proposed method generates fairly good visual quality of images while maintaining almost the same color as the ground truth image. Moreover, the proposed FCN has much higher performance than conventional interpolation methods and generative adversarial network based Pix2Pix. One interesting result is that the proposed FCN produces shadow-free and daylight color images. This result is caused by the fact that the LiDAR sensor data is produced by the light reflection and is, therefore, not affected by sunlight and shadow.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Asymmetric Encoder-Decoder Structured FCN Based LiDAR to Color Image Generation

Kim

Yoo

Park

et al. 2019

Sensors

View full text Add to dashboard Cite

show abstract

“…VeloFCN [23] projected the point cloud onto a 2D map and made 3D bounding boxes prediction by utilising a 2D fully convolutional network (FCN) [24]. RefCNN [25] generated the front view dense reflection map (DRM) by the reflection intensity from LIDAR. The vehicle detection on DRM was performed by YOLOv2 [14].…”

Section: Related Workmentioning

confidence: 99%

RTL3D: real‐time LIDAR‐based 3D object detection with sparse CNN

Lin

Belyaev

et al. 2020

IET Computer Vision

View full text Add to dashboard Cite

LIDAR (light detection and ranging) based real‐time 3D perception is crucial for applications such as autonomous driving. However, most of the convolutional neural network (CNN) based methods are time‐consuming and computation‐intensive. These drawbacks are mainly attributed to the highly variable density of LIDAR point cloud and the complexity of their pipelines. To find a balance between speed and accuracy for 3D object detection from LIDAR, authors propose RTL3D, a computationally efficient Real‐time LIDAR‐based 3D detector. In RTL3D, an effective voxel‐wise feature representation is utilised to organise unstructured point cloud. By employing a sparse feature learning network (SFLN) on voxelised 3D data, RTL3D exploits the sparsity of point cloud and down‐samples 3D data into 2D. Basing on the generated 2D feature map, an optimised dense detection network (DDN) is applied to regress the oriented bounding box without relying on any predefined anchor boxes. The authors also introduce an incremental data augmentation approach which greatly improves the performance of RTL3D. Empirical experiments on public KITTI benchmark demonstrate that RTL3D achieves a competitive performance with state‐of‐the‐art works on 3D detection task. Owning to the simplicity of its single‐stage and anchor‐free design, RTL3D has a real‐time inference speed of 40 FPS.

show abstract

“…As can be seen from Figure 2 a), this sparse LiDAR representation is then up-sampled using the Delaunay Triangulation [40,47], it generates a mesh from the sparse representation. Then the empty pixels are interpolated via the nearest neighbors.…”

Section: Dense Depth Map Generationmentioning

confidence: 99%

Adaptive Multi-Modal Fusion Instance Segmentation for CAEVs in Complex Conditions: Dataset, Framework and Verifications

Peng¹,

Geng²,

Yin³

et al. 2020

Preprint

View full text Add to dashboard Cite

This paper aims to develop an end-to-end sharpening mixture of experts (SMoE) fusion framework to improve the robustness and accuracy of the perception systems for CAEVs in complex illumination and weather conditions. Three original contributions make our work distinctive from the existing relevant literature. First, we introduce the Complex KITTI dataset which consists of 7481 pairs of modified KITTI RGB images and the generated LiDAR dense depth maps, this dataset is fine annotated in instance-level with our proposed semi-automatic annotation method. Second, the SMoE fusion approach is devised to adaptively learn the robust kernels from complementary modalities. Finally, we implement comprehensive comparative experiments, the results show that our proposed SMoE framework yield significant improvements over the other fusion techniques in adverse environmental conditions.

show abstract

Real-Time Deep ConvNet-Based Vehicle Detection Using 3D-LIDAR Reflection Intensity Data

Cited by 18 publications

References 16 publications

Asymmetric Encoder-Decoder Structured FCN Based LiDAR to Color Image Generation

Asymmetric Encoder-Decoder Structured FCN Based LiDAR to Color Image Generation

RTL3D: real‐time LIDAR‐based 3D object detection with sparse CNN

Adaptive Multi-Modal Fusion Instance Segmentation for CAEVs in Complex Conditions: Dataset, Framework and Verifications

Contact Info

Product

Resources

About