Efficient Semantic Segmentation for Visual Bird’s-Eye View Interpretation

Sämann, Timo; Amende, Karl; Milz, Stefan; Witt, Christian; Simón, Martín; Petzold, Johannes

doi:10.1007/978-3-030-01370-7_53

Cited by 12 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ENet [27], one of the most efficient models used a special encoder-decoder structure to highly reduce computational effort. Recently, [39] applied the Channel Pruning method [13] to the ENet to make it more efficient.…”

Section: Semantic Segmentationmentioning

confidence: 99%

See 1 more Smart Citation

Complexer-YOLO: Real-Time 3D Object Detection and Tracking on Semantic Point Clouds

Simón

Amende

Kraus

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

Self Cite

241

View full text Add to dashboard Cite

Accurate detection of 3D objects is a fundamental problem in computer vision and has an enormous impact on autonomous cars, augmented/virtual reality and many applications in robotics. In this work we present a novel fusion of neural network based state-of-the-art 3D detector and visual semantic segmentation in the context of autonomous driving. Additionally, we introduce Scale-Rotation-Translation score (SRTs), a fast and highly parameterizable evaluation metric for comparison of object detections, which speeds up our inference time up to 20% and halves training time. On top, we apply state-of-the-art online multi target feature tracking on the object measurements to further increase accuracy and robustness utilizing temporal information. Our experiments on KITTI show that we achieve same results as state-of-the-art in all related categories, while maintaining the performance and accuracy trade-off and still run in real-time. Furthermore, our model is the first one that fuses visual semantic with 3D object detection.

show abstract

Section: Semantic Segmentationmentioning

confidence: 99%

“…• Visual Class Features: Incorporation of visual pointwise Class-Features generated by fast camera-based Semantic Segmentation [39].…”

Section: Introductionmentioning

confidence: 99%

Complexer-YOLO: Real-Time 3D Object Detection and Tracking on Semantic Point Clouds

Simón

Amende

Kraus

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

Self Cite

241

View full text Add to dashboard Cite

show abstract

“…Inverse Perspective Mapping (IPM). The most straightforward baseline to construct an HD Map is to map semantic segmentation predictions to the bird's-eye view via IPM [51,52]. Although IPM AP@.2 AP@.5 AP@1. mAP AP@.2 AP@.5 AP@1. mAP AP@.2 AP@.5 AP@1. mAP AP@.2 AP@.…”

Section: Baseline Methodsmentioning

confidence: 99%

HDMapNet: An Online HD Map Construction and Evaluation Framework

Li¹,

Wang²,

Wang³

et al. 2021

Preprint

View full text Add to dashboard Cite

High-definition map (HD map) construction is a crucial problem for autonomous driving. This problem typically involves collecting high-quality point clouds, fusing multiple point clouds of the same scene, annotating map elements, and updating maps constantly. This pipeline, however, requires a vast amount of human efforts and resources which limits its scalability. Additionally, traditional HD maps are coupled with centimeter-level accurate localization which is unreliable in many scenarios [1]. In this paper, we argue that online map learning, which dynamically constructs the HD maps based on local sensor observations, is a more scalable way to provide semantic and geometry priors to self-driving vehicles than traditional pre-annotated HD maps. Meanwhile, we introduce an online map learning method, titled HDMapNet. It encodes image features from surrounding cameras and/or point clouds from LiDAR, and predicts vectorized map elements in the bird's-eye view. We benchmark HDMapNet on the nuScenes dataset and show that in all settings, it performs better than baseline methods. Of note, our fusion-based HDMapNet outperforms existing methods by more than 50% in all metrics. To accelerate future research, we develop customized metrics to evaluate map learning performance, including both semantic-level and instancelevel ones. By introducing this method and metrics, we invite the community to study this novel map learning problem. We will release our code and evaluation kit to facilitate future development.

show abstract

“…Relatively few works however have tackled the more specific problem of generating semantic maps from images. Some use the IPM approach mentioned above to map a semantic segmentation of the image plane into the birds-eyeview space [8,23], an approach which works well for estimating local road layout but which fails for objects such as cars and pedestrians which lie above the ground plane. [13] take advantage of RGB-D images to learn an implicit map representation which can be used for later localisation.…”

Section: Related Workmentioning

confidence: 99%

“…We present a simple baseline inspired by other works [8,23] of mapping an image-based semantic segmentation to the ground plane via a homography. The image-level segmentation is computed using a state-of-the-art DeepLabv3 [6] network, pretrained on Cityscapes [7], which shares many classes in common with both NuScenes and Argoverse.…”

Section: Inverse Perspective Mapping (Ipm)mentioning

confidence: 99%

Predicting Semantic Map Representations from Images using Pyramid Occupancy Networks

Roddick¹,

Cipolla²

2020

Preprint

View full text Add to dashboard Cite

Autonomous vehicles commonly rely on highly detailed birds-eye-view maps of their environment, which capture both static elements of the scene such as road layout as well as dynamic elements such as other cars and pedestrians. Generating these map representations on the fly is a complex multi-stage process which incorporates many important vision-based elements, including ground plane estimation, road segmentation and 3D object detection. In this work we present a simple, unified approach for estimating maps directly from monocular images using a single endto-end deep learning architecture. For the maps themselves we adopt a semantic Bayesian occupancy grid framework, allowing us to trivially accumulate information over multiple cameras and timesteps. We demonstrate the effectiveness of our approach by evaluating against several challenging baselines on the NuScenes and Argoverse datasets, and show that we are able to achieve a relative improvement of 9.1% and 22.3% respectively compared to the bestperforming existing method. 1

show abstract

Efficient Semantic Segmentation for Visual Bird’s-Eye View Interpretation

Cited by 12 publications

References 14 publications

Complexer-YOLO: Real-Time 3D Object Detection and Tracking on Semantic Point Clouds

Complexer-YOLO: Real-Time 3D Object Detection and Tracking on Semantic Point Clouds

HDMapNet: An Online HD Map Construction and Evaluation Framework

Predicting Semantic Map Representations from Images using Pyramid Occupancy Networks

Contact Info

Product

Resources

About