Search

Accident anticipation attempts to predict whether an accident may occur in advance, which is greatly significant for improving the safety of intelligent vehicles. Most existing approaches integrate the features of accident-relevant agents with spatial information for accident anticipation. However, these approaches ignore the actual spatio-temporal state of missing agents, whether they are occluded or left, which is relevant to accident occurrence. To address this issue, we propose a Graph and Spatio-temporal Continuity based framework for accident anticipation called GSC, which takes the missing agents into account. Specifically, the proposed GSC maintains the spatio-temporal continuity of missing agents, which are in the occluded spatial state in the process of the graph convolution operation. Besides, we define a novel adjacent matrix to add dynamic features to graph learning. In particular, our adjacent matrix utilizes the historical trajectory of each agent to integrate dynamic information into the original static interaction, which improves the quality of the spatial relation feature for accident anticipation. Experimental results on public datasets demonstrate state-of-the-art performance in the correctness of identifying an accident, which is found to reach over 60%.

Domain generalization (DG) aims to learn a model that generalizes well to an unseen test distribution. Mainstream methods follow the domain-invariant representational learning philosophy to achieve this goal. However, due to the lack of priori knowledge to determine which features are domain specific and task-independent, and which features are domain invariant and task relevant, existing methods typically learn entangled representations, limiting their capacity to generalize to the distribution-shifted target domain. To address this issue, in this article, we propose novel disentangled domain-invariant feature learning networks (D2IFLN) to realize feature disentanglement and facilitate domain-invariant feature learning. Specifically, we introduce a semantic disentanglement network and a domain disentanglement network, disentangling the learned domain-invariant features from both domain-specific class-irrelevant features and domain-discriminative features. To avoid the semantic confusion in adversarial learning for domain-invariant feature learning, we further introduce a graph neural network to aggregate different domain semantic features during model training. Extensive experiments on three DG benchmarks show that the proposed D2IFLN performs better than the state of the art.

Open-set domain adaptation (OSDA) aims to achieve knowledge transfer in the presence of both domain shift and label shift, which assumes that there exist additional unknown target classes not presented in the source domain. To solve the OSDA problem, most existing methods introduce an additional unknown class to the source classifier and represent the unknown target instances as a whole. However, it is unreasonable to treat all unknown target instances as a group since these unknown instances typically consist of distinct categories and distributions. It is challenging to identify all unknown instances with only one additional class. In addition, most existing methods directly introduce marginal distribution alignment to alleviate distribution shift between the source and target domains, failing to learn discriminative class boundaries in the target domain since they ignore categorical discriminative information in the adaptation. To address these problems, in this article, we propose a novel prototype-based shared-dummy classifier (PSDC) model for the OSDA. Specifically, our PSDC introduces an auxiliary dummy classifier to calibrate the source classifier and simultaneously develops a weighted adaptation procedure to align class-wise prototypes for adaptation. We further design a pseudo-unknown learning algorithm to reduce the open-set risk. Extensive experiments on Office-31, Office-Home, and VisDA datasets show that the proposed PSDC can outperform existing methods and achieve the new state-of-the-art performance. The code will be made public.