Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks

Gupta, Agrim; Johnson, Justin; Li, Feifei; Savarese, Silvio; Alahi, Alexandre

doi:10.1109/cvpr.2018.00240

Cited by 1,803 publications

(2,115 citation statements)

References 49 publications

Supporting

Mentioning

2,096

Contrasting

Unclassified

Order By: Relevance

“…Encoder-decoder architectures excel in many sequential problems like trajectory prediction [1], [2]. Encoder-decoder architectures consist of two separate recurrent networks: the encoder processes the input sequence x i = (x i,1 , .…”

Section: Encoder-decodermentioning

confidence: 99%

“…A multitude of models have been proposed for the problem of predicting trajectories, mainly concerning pedestrians [2], [1]. We employ the encoder-decoder model from Section III-C for this task, using the Stanford Drone Dataset (SDD) [22].…”

Section: Trajectory Predictionmentioning

confidence: 99%

“…Additionally, we introduce a new metric tailored for ambiguous problems, as we found previous ones to be unsuited. Indeed, many previous works use some form of oracle metric, which only considers the best hypothesis [1], [2]. Although theoretically sound, this could easily be fooled by guessing a multitude of diverse solutions in hopes of approximating the correct one.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Ambiguity in Sequential Data: Predicting Uncertain Futures With Recurrent Models

Berlati

Scheel

Stefano

et al. 2020

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Ambiguity is inherently present in many machine learning tasks, but especially for sequential models seldom accounted for, as most only output a single prediction. In this work we propose an extension of the Multiple Hypothesis Prediction (MHP) model to handle ambiguous predictions with sequential data, which is of special importance, as often multiple futures are equally likely. Our approach can be applied to the most common recurrent architectures and can be used with any loss function. Additionally, we introduce a novel metric for ambiguous problems, which is better suited to account for uncertainties and coincides with our intuitive understanding of correctness in the presence of multiple labels. We test our method on several experiments and across diverse tasks dealing with time series data, such as trajectory forecasting and maneuver prediction, achieving promising results.

show abstract

Section: Encoder-decodermentioning

confidence: 99%

Section: Trajectory Predictionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Ambiguity in Sequential Data: Predicting Uncertain Futures With Recurrent Models

Berlati

Scheel

Stefano

et al. 2020

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

show abstract

“…If Gaussian noise were used as auxiliary input, an array of Gaussian noise was feed-forwarded together with an MRI slice in the training process as follows: 10 different sets of Gaussian noise were first generated and only the "best" set (i.e., the set that yielded the lowest M * loss (Equation 1)) was used to update the DEP model's parameters. Note that this approach is similar to and inspired by Min-of-N loss in 3D object reconstruction (Fan et al, 2017) and variety loss in Social GAN (Gupta et al, 2018). In the testing process, 10 different sets of Gaussian noise were generated and the average performance was calculated.…”

Section: Experiments Setupmentioning

confidence: 99%

Automatic Spatial Estimation of White Matter Hyperintensities Evolution in Brain MRI using Disease Evolution Predictor Deep Neural Networks

Rachmadi

Valdés-Hernández²,

Makin³

et al. 2019

Preprint

View full text Add to dashboard Cite

AbstractPrevious studies have indicated that white matter hyperintensities (WMH), the main radiological feature of small vessel disease, may evolve (i.e., shrink, grow) or stay stable over a period of time. Predicting these changes are challenging because it involves some unknown clinical risk factors that leads to a non-deterministic prediction task. In this study, we propose a deep learning model to predict the evolution of WMH from baseline to follow-up (i.e., 1-year later), namely “Disease Evolution Predictor” (DEP) model, which can be adjusted to become a non-deterministic model. The DEP model receives a baseline image as input and produces a map called “Disease Evolution Map” (DEM), which represents the evolution of WMH from baseline to follow-up. Two DEP models are proposed, namely DEP-UResNet and DEP-GAN, which are representatives of the supervised (i.e., need expert-generated manual labels to generate the output) and unsupervised (i.e., do not require manual labels produced by experts) deep learning algorithms respectively. To simulate the non-deterministic and unknown parameters involved in WMH evolution, we modulate a Gaussian noise array to the DEP model as auxiliary input. This forces the DEP model to imitate a wider spectrum of alternatives in the prediction results. The alternatives of using other types of auxiliary input instead, such as baseline WMH and stroke lesion loads are also proposed and tested. Based on our experiments, the fully supervised machine learning scheme DEP-UResNet regularly performed better than the DEP-GAN which works in principle without using any expert-generated label (i.e., unsupervised). However, a semi-supervised DEP-GAN model, which uses probability maps produced by a supervised segmentation method in the learning process, yielded similar performances to the DEP-UResNet and performed best in the clinical evaluation. Furthermore, an ablation study showed that an auxiliary input, especially the Gaussian noise, improved the performance of DEP models compared to DEP models that lacked the auxiliary input regardless of the model’s architecture. To the best of our knowledge, this is the first extensive study on modelling WMH evolution using deep learning algorithms, which deals with the non-deterministic nature of WMH evolution.

show abstract

“…It is shown that the proposed neural network was capable of reproducing more realistic flow in multiple scenarios. Besides, other newly developed deep learning method, such as generative adversarial network (GAN) [12,13], also has begun applying on pedestrian trajectory prediction. However, the pedestrian movement modeling researches based on artificial neural network are still insufficient and are necessary to make further studies.…”

Section: Introductionmentioning

confidence: 99%

Artificial neural network based modeling on unidirectional and bidirectional pedestrian flow at straight corridors

Zhao

Long

Zhang

et al. 2020

Physica A: Statistical Mechanics and its Applications

View full text Add to dashboard Cite

Pedestrian modeling is a good way to predict pedestrian movement and thus can be used for controlling pedestrian crowds and guiding evacuations in emergencies. In this paper, we propose a pedestrian movement model based on artificial neural network. In the model, the pedestrian velocity vectors are predicted with two sub models, Semicircular Forward Space Based submodel (SFSB-submodel) and Rectangular Forward Space Based submodel (RFSB-submodel), respectively. Both unidirectional and bidirectional pedestrian flow at straight corridors are investigated by comparing the simulation and the corresponding experimental results. It is shown that the pedestrian trajectories and the fundamental diagrams from the model are all consistent with that from experiments. And the typical lane-formation phenomena are observed in bidirectional flow simulation. In addition, to quantitatively evaluate the precision of the prediction, the mean destination error (MDE) and mean trajectory error (MTE) are defined and calculated to be approximately 0.2m and 0.12m in unidirectional flow scenario. In bidirectional flow, relative distance error (RDE) is about 0.15m. The findings indicate that the proposed model is reasonable and capable of simulating the unidirectional and bidirectional pedestrian flow illustrated in this paper.

show abstract

Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks

Cited by 1,803 publications

References 49 publications

Ambiguity in Sequential Data: Predicting Uncertain Futures With Recurrent Models

Ambiguity in Sequential Data: Predicting Uncertain Futures With Recurrent Models

Automatic Spatial Estimation of White Matter Hyperintensities Evolution in Brain MRI using Disease Evolution Predictor Deep Neural Networks

Artificial neural network based modeling on unidirectional and bidirectional pedestrian flow at straight corridors

Contact Info

Product

Resources

About