Adversarial Geometry-Aware Human Motion Prediction

Gui, Liang-Yan; Wang, Yu-Xiong; Liang, Xiaodan; Moura, J.M.F.

doi:10.1007/978-3-030-01225-0_48

Cited by 200 publications

(159 citation statements)

References 38 publications

Supporting

Mentioning

158

Contrasting

Unclassified

Order By: Relevance

“…Prior works have difficulty in modeling complicated and highly stochastic motions that even the zero-velocity baseline can easily outperform them, as observed by Martinez et al [6]. The adversarial loss in AGED [8] also leads to a significant performance improvement in aperiodic motions, but here we outperform them all, as can be seen in the tasks greeting, sitting and taking photo (see Table II). Fig.…”

Section: Short-term Motion Predictionmentioning

confidence: 50%

“…Our method sometimes has difficulty in the first few output time-steps. This is expected since we are adding the residual angle to reconstructed poses rather than the last pose from the original input sequence as done in [6], [8]. However, our method excels for longer temporal horizons.…”

Section: Short-term Motion Predictionmentioning

confidence: 96%

“…We use two normalization methods depending on the baseline that we are comparing against: (1) subtract the mean and normalized between -1 and 1, which is used in [11], and (2) subtract the mean and divide by the standard deviation, which is used by the other methods [6], [8], [21]. Table I: Comparison of mean angle error between our method and top performing baselines for short-term motion prediction.…”

Section: B Data Preprocessingmentioning

confidence: 99%

See 2 more Smart Citations

Human Motion Prediction Via Pattern Completion in Latent Representation Space

Meger

2019

2019 16th Conference on Computer and Robot Vision (CRV)

View full text Add to dashboard Cite

Inspired by ideas in cognitive science, we propose a novel and general approach to solve human motion understanding via pattern completion on a learned latent representation space. Our model outperforms current state-of-the-art methods in human motion prediction across a number of tasks, with no customization. To construct a latent representation for time-series of various lengths, we propose a new and generic autoencoder based on sequence-to-sequence learning. While traditional inference strategies find a correlation between an input and an output, we use pattern completion, which views the input as a partial pattern and to predict the best corresponding complete pattern. Our results demonstrate that this approach has advantages when combined with our autoencoder in solving human motion prediction, motion generation and action classification.

show abstract

Section: Short-term Motion Predictionmentioning

confidence: 50%

Section: Short-term Motion Predictionmentioning

confidence: 96%

Section: B Data Preprocessingmentioning

confidence: 99%

See 1 more Smart Citation

Human Motion Prediction Via Pattern Completion in Latent Representation Space

Meger

2019

2019 16th Conference on Computer and Robot Vision (CRV)

View full text Add to dashboard Cite

show abstract

“…Neural generative modeling addresses this problem, including Generative Adversarial Networks (Mathieu et al, 2016;Luc et al, 2017) and Variational Auto-Encoders (Walker et al, 2016). Both GANs (Villegas et al, 2017;Kiasari et al, 2018;Gui et al, 2018;Lin and Amer, 2018;Wang et al, 2018) and VAEs (Walker et al, 2017;Bütepage et al, 2018) have been applied to the task of human motion prediction. A recent work, (Gui et al, 2018), is of particular interest, as it shows strong performance by proposing two distinct discriminators learned jointly with the sequence generator.…”

Section: Learning a Stochastic Processmentioning

confidence: 99%

“…(Martinez et al, CVPR 2017) 0.28 0.49 0.72 0.81 0.23 0.39 0.62 0.76 0.33 0.61 1.05 1.15 0.31 0.68 1.01 1.09 Adversarial (Gui et al, ECCV 2018) 0 erence implementation samples only four chunks from each test sequence at random positions, using a fixed seed to initialize the random generator 1 . This exact methodology is adopted by Liu et al (2016); Martinez et al (2017); Pavllo et al (2018b); Gui et al (2018) and makes the quantitative results across these papers comparable.…”

Section: More Consistent Short-term Evaluationmentioning

confidence: 99%

Modeling Human Motion with Quaternion-Based Neural Networks

et al. 2019

View full text Add to dashboard Cite

Previous work on predicting or generating 3D human pose sequences regresses either joint rotations or joint positions. The former strategy is prone to error accumulation along the kinematic chain, as well as discontinuities when using Euler angles or exponential maps as parameterizations. The latter requires re-projection onto skeleton constraints to avoid bone stretching and invalid configurations. This work addresses both limitations. QuaterNet represents rotations with quaternions and our loss function performs forward kinematics on a skeleton to penalize absolute position errors instead of angle errors. We investigate both recurrent and convolutional architectures and evaluate on short-term prediction and long-term generation. For the latter, our approach is qualitatively judged as realistic as recent neural strategies from the graphics literature. Our experiments compare quaternions to Euler angles as well as exponential maps and show that only a very short context is required to make reliable future predictions. Finally, we show that the standard evaluation protocol for Human3.6M produces high variance results and we propose a simple solution.

show abstract

Generative model‐enhanced human motion prediction

et al. 2022

View full text Add to dashboard Cite

The task of predicting human motion is complicated by the natural heterogeneity and compositionality of actions, necessitating robustness to distributional shifts as far as out-of-distribution (OoD). Here, we formulate a new OoD benchmark based on the Human3.6M and Carnegie Mellon University (CMU) motion capture datasets, and introduce a hybrid framework for hardening discriminative architectures to OoD failure by augmenting them with a generative model. When applied to current state-of-the-art discriminative models, we show that the proposed approach improves OoD robustness without sacrificing in-distribution performance, and can theoretically facilitate model interpretability. We suggest human motion predictors ought to be constructed with OoD challenges in mind, and provide an extensible general framework for hardening diverse discriminative architectures to extreme distributional shift.

show abstract

Adversarial Geometry-Aware Human Motion Prediction

Cited by 200 publications

References 38 publications

Human Motion Prediction Via Pattern Completion in Latent Representation Space

Human Motion Prediction Via Pattern Completion in Latent Representation Space

Modeling Human Motion with Quaternion-Based Neural Networks

Generative model‐enhanced human motion prediction

Contact Info

Product

Resources

About