Reasoning about the motions of other objects is important for moving platforms such as intelligent vehicles. In this paper, we propose a pedestrian trajectory prediction approach that utilizes pedestrian appearance as well as historical locations and camera ego-motion to predict future trajectories of pedestrians for vehicles. The proposed model consists of three encoders and a decoder, namely: appearance encoder, location encoder, and ego-motion encoder; and a Long Short-term Memory (LSTM) decoder. The appearance encoder employs an embedding Convolutional Neural Network (CNN) and a Convolutional LSTM (ConvLSTM) to encode pedestrian appearance. A disparity-guided attention mechanism is designed in the appearance encoder to attend to salient dynamic regions in pedestrian appearance. The location encoder and ego-motion encoder employ LSTMs to encode pedestrian historical locations and camera ego-motion. The outputs of all three encoders are concatenated as a representation of input information, which is decoded by the decoder to finally generate the future trajectories of observed pedestrians. We evaluated the proposed model on two public datasets with several baselines and the results showed improvement of about 20% compared with the baselines. We also visualized the predicted trajectories and heat maps generated by disparity-guided attention. Both the visualization and quantitative results validated that our proposed model outperforms the baselines on predicting accuracy and that the model is feasible to capture salient spatiotemporal regions in pedestrian appearance.

LiDAR point clouds are significantly impacted by snow in driving scenarios, introducing scattered noise points and phantom objects, thereby compromising the perception capabilities of autonomous driving systems. Current effective methods for removing snow from point clouds largely rely on outlier filters, which mechanically eliminate isolated points. This research proposes a novel translation model for LiDAR point clouds, the ‘L-DIG’ (LiDAR depth images GAN), built upon refined generative adversarial networks (GANs). This model not only has the capacity to reduce snow noise from point clouds, but it also can artificially synthesize snow points onto clear data. The model is trained using depth image representations of point clouds derived from unpaired datasets, complemented by customized loss functions for depth images to ensure scale and structure consistencies. To amplify the efficacy of snow capture, particularly in the region surrounding the ego vehicle, we have developed a pixel-attention discriminator that operates without downsampling convolutional layers. Concurrently, the other discriminator equipped with two-step downsampling convolutional layers has been engineered to effectively handle snow clusters. This dual-discriminator approach ensures robust and comprehensive performance in tackling diverse snow conditions. The proposed model displays a superior ability to capture snow and object features within LiDAR point clouds. A 3D clustering algorithm is employed to adaptively evaluate different levels of snow conditions, including scattered snowfall and snow swirls. Experimental findings demonstrate an evident de-snowing effect, and the ability to synthesize snow effects.

Pedestrian action prediction is crucial for many applications such as autonomous driving. However, state-of-the-art methods lack the explainability needed for trustworthy predictions. In this paper, a novel framework called MulCPred is proposed that explains its predictions based on multi-modal concepts represented by training samples. Previous concept-based methods have limitations, including the following: (1) they cannot be directly applied to multi-modal cases; (2) they lack the locality needed to attend to details in the inputs; (3) they are susceptible to mode collapse. These limitations are tackled accordingly through the following approaches: (1) a linear aggregator to integrate the activation results of the concepts into predictions, which associates concepts of different modalities and provides ante hoc explanations of the relevance between the concepts and the predictions; (2) a channel-wise recalibration module that attends to local spatiotemporal regions, which enables the concepts with locality; (3) a feature regularization loss that encourages the concepts to learn diverse patterns. MulCPred is evaluated on multiple datasets and tasks. Both qualitative and quantitative results demonstrate that MulCPred is promising in improving the explainability of pedestrian action prediction without obvious performance degradation. Moreover, by removing unrecognizable concepts, MulCPred shows improved cross-dataset prediction performance, suggesting its potential for further generalization.

Predicting pedestrian behavior is challenging yet crucial for applications such as autonomous driving and smart cities. Recent deep learning models have achieved remarkable performance in making accurate predictions, but they fail to provide explanations of their inner workings. One reason for this problem is the multi-modal inputs. To bridge this gap, we present Sparse Prototype Network (SPN), an explainable method designed to simultaneously predict a pedestrian's future action, trajectory, and pose. SPN leverages an intermediate prototype bottleneck layer to provide sample-based explanations for its predictions. The prototypes are modality-independent, meaning that they can correspond to any modality from the input. Therefore, SPN can extend to arbitrary combinations of modalities. Regularized by mono-semanticity and clustering constraints, the prototypes learn consistent and human-understandable features and achieve state-of-the-art performance on action, trajectory and pose prediction on TITAN and PIE. Finally, we propose a metric named Top-K Mono-semanticity Scale to quantitatively evaluate the explainability. Qualitative results show a positive correlation between sparsity and explainability.

Pedestrian action prediction is of great significance for many applications such as autonomous driving. However, state-of-the-art methods lack explainability to make trustworthy predictions. In this paper, a novel framework called MulCPred is proposed that explains its predictions based on multi-modal concepts represented by training samples. Previous concept-based methods have limitations including: 1) they cannot directly apply to multi-modal cases; 2) they lack locality to attend to details in the inputs; 3) they suffer from mode collapse. These limitations are tackled accordingly through the following approaches: 1) a linear aggregator to integrate the activation results of the concepts into predictions, which associates concepts of different modalities and provides ante-hoc explanations of the relevance between the concepts and the predictions; 2) a channel-wise recalibration module that attends to local spatiotemporal regions, which enables the concepts with locality; 3) a feature regularization loss that encourages the concepts to learn diverse patterns. MulCPred is evaluated on multiple datasets and tasks. Both qualitative and quantitative results demonstrate that MulCPred is promising in improving the explainability of pedestrian action prediction without obvious performance degradation. Furthermore, by removing unrecognizable concepts from MulCPred, the cross-dataset prediction performance is improved, indicating the feasibility of further generalizability of MulCPred.