Complex-YOLO: Real-time 3D Object Detection on Point Clouds

By Martin Simon, Stefan Milz, Karl Amende, Horst-Michael Groß. 2018

165 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, 3D Vision & Pose

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

In 3D Vision & Pose

Citations in the corpus

WoodScape: A Multi-Task, Multi-Camera Fisheye Dataset for Autonomous Driving (2019, methodology): “For baseline, we trained Complex-YOLO [50] for a single class (‘car’).”
YOLO3D: End-to-end real-time 3D Oriented Object Bounding Box Detection from LiDAR Point Cloud (2018, methodology): “Extending YOLOv2 [3], we include the orientation of the OBB as a direct regression task, unlike the work in [14], which suggests two separate losses for the real and imaginary parts of the angle, without explicit nor implicit correlation between them, which may result in wrong or invalid angles in m”
PointPillars: Fast Encoders for Object Detection From Point Clouds (2018, methodology): “A common workaround to this problem is to partition the ground plane into a regular grid, for example 10 x 10 cm, and then perform a hand-crafted feature encoding method on the points in each grid cell [2, 11, 26, 32].”
Fast Point R-CNN (2019, background): “One line of solutions is to use a coarse grid [43, 40, 23, 2, 31, 1]; but coarse quantization prevents following CNN from utilizing fine-grained information.”
IntentNet: Learning to Predict Intention from Raw Sensor Data (2018, background): “However, these methods suffer from either limited performance or heavy computation and thus are not suitable for self driving [18].”
SECOND: Sparsely Embedded Convolutional Detection (2018, methodology): “ComplexYOLO [19] uses a YOLO (You Only Look Once) [20] network and a complex angle encoding approach to increase speed and orientation performance, but it uses fixed heights and z-locations in the predicted 3D bounding boxes.”
A Survey on 3D Object Detection Methods for Autonomous Driving Applications (2019, methodology): “This allows Complex-YOLO to achieve a runtime of 50 fps, up to five times more efficient than previous methods, despite inferior, but comparable detection performance.”
Object Detection Algorithm Based on Improved YOLOv3 (2020, methodology): “have developed a complex-YOLO method [17], which uses a specific complex regression strategy to estimate multi-class 3D boxes in Cartesian space for detecting RGB images; the authors report a significant improvement in the speed of 3D object detection.”
PointPainting: Sequential Fusion for 3D Object Detection (2019, methodology): “For methods using hand-coded features [29, 21], some extra work is required to modify the feature encoder.”