Search for: All records

The study of complex human interactions and group activities has become a focal point in human-centric computer vision. However, progress in related tasks is often hindered by the challenges of obtaining large-scale labeled datasets from real-world scenarios. To address the limitation, we introduce M3Act, a synthetic data generator for multi-view multi-group multi-person human atomic actions and group activities. Powered by Unity Engine, M3Act features multiple semantic groups, highly diverse and photorealistic images, and a comprehensive set of annotations, which facilitates the learning of human-centered tasks across singleperson, multi-person, and multi-group conditions. We demonstrate the advantages of M3Act across three core experiments. The results suggest our synthetic dataset can significantly improve the performance of several downstream methods and replace real-world datasets to reduce cost. Notably, M3Act improves the state-of-the-art MOTRv2 on DanceTrack dataset, leading to a hop on the leaderboard from 10th to 2nd place. Moreover, M3Act opens new research for controllable 3D group activity generation. We define multiple metrics and propose a competitive baseline for the novel task. Our code and data are available at our project page: http://cjerry1243.github.io/M3Act. 

FSS (Few-shot segmentation) aims to segment a target class using a small number of labeled images (support set). To extract information relevant to the target class, a dominant approach in best performing FSS methods removes background features using a support mask. We observe that this feature excision through a limiting support mask introduces an information bottleneck in several challenging FSS cases, e.g., for small targets and/or inaccurate target boundaries. To this end, we present a novel method (MSI), which maximizes the support-set information by exploiting two complementary sources of features to generate super correlation maps. We validate the effectiveness of our approach by instantiating it into three recent and strong FSS methods. Experimental results on several publicly available FSS benchmarks show that our proposed method consistently improves performance by visible margins and leads to faster convergence. Our code and trained models are available at: https://github.com/moonsh/MSI-Maximize-Support-Set-Information 

Computational emotion, is naturally predicated on an operating theory of emotion. This paper seeks to explore the prevalence of three different approaches in the literature, namely basic emotion, dimensional emotion, and constructed emotion. Basic emotion maintains that there exists a discrete set of primitive emotions evolved as responses to certain stimuli; dimensional emotion sees different emotions as systematically related by two or more dimensions (typically valence and arousal); and constructed emotion describes emotional experience as a function of the brain’s general predictive faculties applied to learned social concepts of different emotions. In order to see how these approaches are represented in affective computing literature, we conduct a systematic survey spanning the IEEE, ACM, ScienceDirect, and Engineering Village databases. Out of 204 selected papers, 151 apply basic emotion theory, 48 apply dimensional emotion, and 5 apply constructed emotion. We find promising representation of the constructed emotion theory in the affective computing literature and conclude that it provides a theoretical basis worth pursuing for affective engagement human computer interaction (HCI) applications. 

Predicting the crowd behavior in complex environments is a key requirement for crowd and disaster management, architectural design, and urban planning. Given a crowd’s immediate state, current approaches must be successively repeated over multiple time-steps for long-term predictions, leading to compute expensive and error-prone results. However, most applications require the ability to accurately predict hundreds of possible simulation outcomes (e.g., under different environment and crowd situations) at real-time rates, for which these approaches are prohibitively expensive. We propose the first deep framework to instantly predict the long-term flow of crowds in arbitrarily large, realistic environments. Central to our approach are a novel representation CAGE, which efficiently encodes crowd scenarios into compact, fixed-size representations that losslessly represent the environment, and a modified SegNet architecture for instant long-term crowd flow prediction. We conduct comprehensive experiments on novel synthetic and real datasets. Our results indicate that our approach is able to capture the essence of real crowd movement over very long time periods, while generalizing to never-before-seen environments and crowd contexts. The associated Supplementary Material, models, and datasets are available at github.com/SSSohn/LTCF.