All together now: Simultaneous Detection and Continuous Pose Estimation using a Hough Forest with Probabilistic Locally Enhanced Voting

Redondo-Cabrera, Carolina; López-Sastre, Roberto J.; Tuytelaars, Tinne

doi:10.5244/c.28.63

Cited by 21 publications

(38 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The works which are closest to our method are [20] and [25]. [20] relies on a modified Hough regression forest that estimates the pose distribution at each frame, and then fuses it with a smoothed average of the past n distributions before determining the maximum a posteriori pose.…”

Section: Related Workmentioning

confidence: 99%

“…[20] relies on a modified Hough regression forest that estimates the pose distribution at each frame, and then fuses it with a smoothed average of the past n distributions before determining the maximum a posteriori pose. This approach can be prone to error accumulation due to a sequence of wrongly estimated distributions.…”

Section: Related Workmentioning

confidence: 99%

“…Some approaches rely on 3D-based pose estimators trained using CAD models and possibly deliver a real-valued pose [25,18,17,27]. Since the availability of 3D models cannot be guaranteed, others have successfully explored the possibility of relying only on 2D data [20,5]. Regardless of the pose estimator used, the combination in the temporal domain of singleframe results is in most cases obtained through suboptimal strategies, such as aggregation and fusion of pose estimation pdf's over consecutive frames, or computationally ex-pensive techniques, such as particle filtering.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Pose Estimation of Object Categories in Videos Using Linear Programming

Fenzi

Leal-Taixé

Schindler

et al. 2015

2015 IEEE Winter Conference on Applications of Computer Vision

View full text Add to dashboard Cite

In this paper we propose a method to consistently recover the pose of an object from a known class in a video sequence. As individual poses estimated from monocular images are rather noisy, we optimally aggregate pose evidence over all video frames. We construct a graph where nodes are values sampled from the pose posterior distributions computed by a continuous pose estimator in each frame of the sequence. We then find the globally optimum pose path through the graph that best explains the pose evidence for the whole sequence. As a result, we recover the correct object orientation at each frame even if single-frame pose evidence is sometimes inaccurate. We evaluate our approach on two publicly available car datasets, which encompass busy street scenarios and car races with significant changes in car orientation, blur and occlusions. We show that our method outperforms state-of-the-art approaches reducing the error by 40% on the challenging KITTI dataset.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Pose Estimation of Object Categories in Videos Using Linear Programming

Fenzi

Leal-Taixé

Schindler

et al. 2015

2015 IEEE Winter Conference on Applications of Computer Vision

View full text Add to dashboard Cite

show abstract

“…A number of methods have been proposed to detect and track people in depth images [25,18], particularly those generated using sensors based on structured light projection, such as the first version of Kinect. Although the full pipeline implemented in Kinect2 SDK has not been published, we have performed a set of preliminary experiments comparing this method with other state of the art implementations available for 3D head tracking from depth measurements, such as the method of Fanelli et al [6] and RGB methods, such as that of [20].…”

Section: Depth-based Likelihood Functionmentioning

confidence: 99%

Person Tracking Using Audio and Depth Cues

Liu

Campos

Wang

et al. 2015

2015 IEEE International Conference on Computer Vision Workshop (ICCVW)

View full text Add to dashboard Cite

In this paper, a novel probabilistic Bayesian tracking scheme is proposed and applied to bimodal measurements consisting of tracking results from the depth sensor and audio recordings collected using binaural microphones. We use random finite sets to cope with varying number of tracking targets. A measurement-driven birth process is integrated to quickly localize any emerging person. A new bimodal fusion method that prioritizes the most confident modality is employed. The approach was tested on real room recordings and experimental results show that the proposed combination of audio and depth outperforms individual modalities, particularly when there are multiple people talking simultaneously and when occlusions are frequent.

show abstract

“…A number of methods have been proposed to detect and track people in depth images [12,13], particularly those genWe are grateful for the help of Mark Barnard on building the dataset used in this paper. We would like to acknowledge the support of the EPSRC Programme Grant S3A: Future Spatial Audio for an Immersive Listener Experience at Home (EP/L000539/1) and the BBC as part of the BBC Audio Research Partnership.…”

Section: Introductionmentioning

confidence: 99%

Identity association using PHD filters in multiple head tracking with depth sensors

Liu

Campos

Wang

et al. 2016

2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

The work on 3D human pose estimation has seen a significant amount of progress in recent years, particularly due to the widespread availability of commodity depth sensors. However, most pose estimation methods follow a trackingas-detection approach which does not explicitly handle occlusions, thus introducing outliers and identity association issues when multiple targets are involved. To address these issues, we propose a new method based on Probability Hypothesis Density (PHD) filter. In this method, the PHD filter with a novel clutter intensity model is used to remove outliers in the 3D head detection results, followed by an identity association scheme with occlusion detection for the targets. Experimental results show that our proposed method greatly mitigates the outliers, and correctly associates identities to individual detections with low computational cost.

show abstract

All together now: Simultaneous Detection and Continuous Pose Estimation using a Hough Forest with Probabilistic Locally Enhanced Voting

Cited by 21 publications

References 19 publications

Pose Estimation of Object Categories in Videos Using Linear Programming

Pose Estimation of Object Categories in Videos Using Linear Programming

Person Tracking Using Audio and Depth Cues

Identity association using PHD filters in multiple head tracking with depth sensors

Contact Info

Product

Resources

About