Goto

Collaborating Authors

Learning person-object interactions for action recognition in still images

Neural Information Processing Systems

We investigate a discriminatively trained model of person-object interactions for recognizing common human actions in still images. We build on the locally order-less spatial pyramid bag-of-features model, which was shown to perform extremely well on a range of object, scene and human action recognition tasks. We introduce three principal contributions. First, we replace the standard quantized local HOG/SIFT features with stronger discriminatively trained body part and object detectors. Second, we introduce new person-object interaction features based on spatial co-occurrences of individual body parts and objects. Third, we address the combinatorial problem of a large number of possible interaction pairs and propose a discriminative selection procedure using a linear support vector machine (SVM) with a sparsity inducing regularizer. Learning of action-specific body part and object interactions bypasses the difficult problem of estimating the complete human body pose configuration. Benefits of the proposed model are shown on human action recognition in consumer photographs, outperforming the strong bag-of-features baseline.


Responsive Planning and Recognition for Closed-Loop Interaction

arXiv.org Artificial Intelligence

Many intelligent systems currently interact with others using at least one of fixed communication inputs or preset responses, resulting in rigid interaction experiences and extensive efforts developing a variety of scenarios for the system. Fixed inputs limit the natural behavior of the user in order to effectively communicate, and preset responses prevent the system from adapting to the current situation unless it was specifically implemented. Closed-loop interaction instead focuses on dynamic responses that account for what the user is currently doing based on interpretations of their perceived activity. Agents employing closed-loop interaction can also monitor their interactions to ensure that the user responds as expected. We introduce a closed-loop interactive agent framework that integrates planning and recognition to predict what the user is trying to accomplish and autonomously decide on actions to take in response to these predictions. Based on a recent demonstration of such an assistive interactive agent in a turn-based simulated game, we also discuss new research challenges that are not present in the areas of artificial intelligence planning or recognition alone.


Structural mechanisms of centromeric nucleosome recognition by the kinetochore protein CENP-N

Science

Accurate chromosome segregation requires the proper assembly of kinetochore proteins. A key step in this process is the recognition of the histone H3 variant CENP-A in the centromeric nucleosome by the kinetochore protein CENP-N. We report cryo–electron microscopy (cryo-EM), biophysical, biochemical, and cell biological studies of the interaction between the CENP-A nucleosome and CENP-N. We show that human CENP-N confers binding specificity through interactions with the L1 loop of CENP-A, stabilized by electrostatic interactions with the nucleosomal DNA. Mutational analyses demonstrate analogous interactions in Xenopus, which are further supported by residue-swapping experiments involving the L1 loop of CENP-A.


[Review] Mechanisms for initiating cellular DNA replication

Science

Early work in bacterial systems established that initiator proteins recognize specific sequence elements (termed replication origins). Nucleotide-dependent oligomerization of the bacterial initiator DnaA at origins facilitates DNA melting and provides an access point for the replicative helicase, DnaB, which is recruited to and loaded onto each of the single-stranded origin regions by interactions with the initiator and other loader proteins to establish bidirectional replication forks. It has recently become clear that the strategies for origin recognition, helicase loading, and duplex DNA melting in archaeal and eukaryal systems, as well as the order of these steps, deviate considerably from the archetypal initiation events followed by bacteria. Although nucleotide-dependent interactions between initiator subunits still control early steps of replication initiation and regulate the association of initiators with origin DNA, they do not appear to contribute to DNA melting. Instead, archaeal/eukaryotic initiators [Orc in archaea or the multisubunit origin recognition complex (ORC) in eukaryotes] and helicase coloading factors cooperate to recruit and load the replicative helicase motor, the minichromosome maintenance (MCM) complex, onto duplex DNA in an inactive, double-hexameric form.


Learning person-object interactions for action recognition in still images

Neural Information Processing Systems

We investigate a discriminatively trained model of person-object interactions for recognizing common human actions in still images. We build on the locally order-less spatial pyramid bag-of-features model, which was shown to perform extremely well on a range of object, scene and human action recognition tasks. We introduce three principal contributions. First, we replace the standard quantized local HOG/SIFT features with stronger discriminatively trained body part and object detectors. Second, we introduce new person-object interaction features based on spatial co-occurrences of individual body parts and objects.