Human Pose Estimation in Real Traffic Scenes

Kress, Viktor; Jung, Janis; Zernetsch, Stefan; Doll, Konrad; Sick, Bernhard

doi:10.1109/ssci.2018.8628660

Cited by 12 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zernetsch et al [13] developed a probabilistic VRU trajectory forecasting method. Kress et al [14] used this sensor setup as a reference to evaluate a human keypoint detection model deployed to a mobile research vehicle. It is worth mentioning that this sensor setup and the knowledge from the Ko-PER and DeCoInt 2 projects were utilized to develop the novel proposed sensor setup.…”

Section: A Intelligent Intersectionsmentioning

confidence: 99%

See 1 more Smart Citation

The IMPTC Dataset: An Infrastructural Multi-Person Trajectory and Context Dataset

Hetzel

Reichert

Reitberger

et al. 2023

2023 IEEE Intelligent Vehicles Symposium (IV)

Self Cite

View full text Add to dashboard Cite

Inner-city intersections are among the most critical traffic areas for injury and fatal accidents. Automated vehicles struggle with the complex and hectic everyday life within those areas. Sensor-equipped smart infrastructures, which can cooperate with vehicles, can benefit automated traffic by extending the perception capabilities of drivers and vehicle perception systems. Additionally, they offer the opportunity to gather reproducible and precise data of a holistic scene understanding, including context information as a basis for training algorithms for various applications in automated traffic. Therefore, we introduce the Infrastructural Multi-Person Trajectory and Context Dataset (IMPTC). We use an intelligent public inner-city intersection in Germany with visual sensor technology. A multi-view camera and LiDAR system perceives traffic situations and road users' behavior. Additional sensors monitor contextual information like weather, lighting, and traffic light signal status. The data acquisition system focuses on Vulnerable Road Users (VRUs) and multiagent interaction. The resulting dataset consists of eight hours of measurement data. It contains over 2,500 VRU trajectories, including pedestrians, cyclists, e-scooter riders, strollers, and wheelchair users, and over 20,000 vehicle trajectories at different day times, weather conditions, and seasons. In addition, to enable the entire stack of research capabilities, the dataset includes all data, starting from the sensor-, calibration-and detection data until trajectory and context data. The dataset is continuously expanded and is available online for non-commercial research at https://github.com/kav-institute/imptc-dataset.

show abstract

Section: A Intelligent Intersectionsmentioning

confidence: 99%

“…The model's outcome is used with trajectory data as additional information for VRU trajectory prediction. Kress et al [14] demonstrated that human bodypose information helps to improve VRU intention detection. Instead of using one real-world coordinate representing a VRU's 3D location, 17 body joints are used.…”

Section: B Vru Trajectory Datasetsmentioning

confidence: 99%

The IMPTC Dataset: An Infrastructural Multi-Person Trajectory and Context Dataset

Hetzel

Reichert

Reitberger

et al. 2023

2023 IEEE Intelligent Vehicles Symposium (IV)

Self Cite

View full text Add to dashboard Cite

show abstract

“…To obtain the 3D poses of the VRUs, first, the 2D poses, i.e., the two-dimensional coordinates of several joints in the images, were estimated using the CNN proposed in [19] followed by a reconstruction of plausible 3D poses using the approach from [20]. More details on this procedure and an evaluation of its accuracy can be found in [21]. Compared to 2D poses or other image-based methods, 3D poses have the advantage of being independent of the perspective of the recording camera and allow for a compensation of the vehicle's own motion.…”

Section: Datasetmentioning

confidence: 99%

Pose and Semantic Map Based Probabilistic Forecast of Vulnerable Road Users' Trajectories

Kress¹,

Jeske²,

Zernetsch³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

In this article, an approach for probabilistic trajectory forecasting of vulnerable road users (VRUs) is presented, which considers past movements and the surrounding scene. Past movements are represented by 3D poses reflecting the posture and movements of individual body parts. The surrounding scene is modeled in the form of semantic maps showing, e.g., the course of streets, sidewalks, and the occurrence of obstacles. The forecasts are generated in grids discretizing the space and in the form of arbitrary discrete probability distributions. The distributions are evaluated in terms of their reliability, sharpness, and positional accuracy. We compare our method with an approach that provides forecasts in the form of Gaussian distributions and discuss the respective advantages and disadvantages. Thereby, we investigate the impact of using poses and semantic maps. With a technique called spatial label smoothing, our approach achieves reliable forecasts. Overall, the poses have a positive impact on the forecasts. The semantic maps offer the opportunity to adapt the probability distributions to the individual situation, although at the considered forecasted time horizon of 2.52 s they play a minor role compared to the past movements of the VRU. Our method is evaluated on a dataset recorded in inner-city traffic using a research vehicle. The dataset is made publicly available.

show abstract

“…1, articulated pose can be utilized to generate an appearance-invariant intermediate representation and is widely supported in literature for modeling VRU behavior like action recognition (Hariyono and Jo 2015), crossing intention estimation (Fang and López 2019), trajectory prediction (Rasouli et al 2019) and gesture recognition (Tripathi et al 2019). Even though there have been great strides in human pose estimation, there has been less focus from the perspective of real-time applications such as AD usecases (Kothari et al 2017) (Kress et al 2018). Most renowned approaches are either computation-intensive (He et al 2017) or memory-intensive (Huang, Zhu, and Huang 2019), making them today not optimal for real-time AD applications which demands inference on high resolution images with high frame rates.…”

Section: Introductionmentioning

confidence: 99%

VRU Pose-SSD: Multiperson Pose Estimation For Automated Driving

Kumar¹,

Ramesh²,

Chakraborty³

et al. 2021

AAAI

View full text Add to dashboard Cite

We present a fast and efficient approach for joint person detection and pose estimation optimized for automated driving (AD) in urban scenarios. We use a multitask weight sharing architecture to jointly train detection and pose estimation. This modular architecture allows us to accommodate different downstream tasks in the future. By systematic large-scale experiments on the Tsinghua-Daimler Urban Pose Dataset (TDUP), we obtain multiple models with varying accuracy-speed trade-offs. We then quantize and optimize our network for deployment and present a detailed analysis of the efficacy of the algorithm. We introduce a two-stage evaluation strategy, which is more suitable for AD and achieve a significant performance improvement in comparison to state-of-the-art approaches. Our optimized model runs at 52~fps on full HD images and still reaches a competitive performance of 32.25~LAMR. We are confident that our work serves as an enabler to tackle higher-level tasks like VRU intention estimation and gesture recognition, which rely on stable pose estimates and will play a crucial role in future AD systems.

show abstract

Human Pose Estimation in Real Traffic Scenes

Cited by 12 publications

References 21 publications

The IMPTC Dataset: An Infrastructural Multi-Person Trajectory and Context Dataset

The IMPTC Dataset: An Infrastructural Multi-Person Trajectory and Context Dataset

Pose and Semantic Map Based Probabilistic Forecast of Vulnerable Road Users' Trajectories

VRU Pose-SSD: Multiperson Pose Estimation For Automated Driving

Contact Info

Product

Resources

About