Abstract. Much recent research in human activity recognition has focused on the problem of recognizing simple repetitive (walking, running, waving) and punctual actions (sitting up, opening a door, hugging). However, many interesting human activities are characterized by a complex temporal composition of simple actions. Automatic recognition of such complex actions can benefit from a good understanding of the temporal structures. We present in this paper a framework for modeling motion by exploiting the temporal structure of the human activities. In our framework, we represent activities as temporal compositions of motion segments. We train a discriminative model that encodes a temporal decomposition of video sequences, and appearance models for each motion segment. In recognition, a query video is matched to the model according to the learned appearances and motion segment decomposition. Classification is made based on the quality of matching between the motion segment classifiers and the temporal segments in the query sequence. To validate our approach, we introduce a new dataset of complex Olympic Sports activities. We show that our algorithm performs better than other state of the art methods.

Signal transducer and activator of transcription 3 (STAT3) is phosphorylated by various kinases, several of which have been implicated in aberrant fibroblast activation in fibrotic diseases including systemic sclerosis (SSc). Here we show that profibrotic signals converge on STAT3 and that STAT3 may be an important molecular checkpoint for tissue fibrosis. STAT3 signaling is hyperactivated in SSc in a TGFβ-dependent manner. Expression profiling and functional studies in vitro and in vivo demonstrate that STAT3 activation is mediated by the combined action of JAK, SRC, c-ABL, and JNK kinases. STAT3-deficient fibroblasts are less sensitive to the pro-fibrotic effects of TGFβ. Fibroblast-specific knockout of STAT3, or its pharmacological inhibition, ameliorate skin fibrosis in experimental mouse models. STAT3 thus integrates several profibrotic signals and might be a core mediator of fibrosis. Considering that several STAT3 inhibitors are currently tested in clinical trials, STAT3 might be a candidate for molecular targeted therapies of SSc.

Kinesin-3 motor UNC-104/KIF1A is essential for transporting synaptic precursors to synapses. Although the mechanism of cargo binding is well understood, little is known how motor activity is regulated. We mapped functional interaction domains between SYD-2 and UNC-104 by using yeast 2-hybrid and pull-down assays and by using FRET/ fluorescence lifetime imaging microscopy to image the binding of SYD-2 to UNC-104 in living Caenorhabditis elegans. We found that UNC-104 forms SYD-2-dependent axonal clusters (appearing during the transition from L2 to L3 larval stages), which behave in FRAP experiments as dynamic aggregates. High-resolution microscopy reveals that these clusters contain UNC-104 and synaptic precursors (synaptobrevin-1). Analysis of motor motility indicates bi-directional movement of UNC-104, whereas in syd-2 mutants, loss of SYD-2 binding reduces net anterograde movement and velocity (similar after deleting UNC-104's liprin-binding domain), switching to retrograde transport characteristics when no role of SYD-2 on dynein and conventional kinesin UNC-116 motility was found. These data present a kinesin scaffolding protein that controls both motor clustering along axons and motor motility, resulting in reduced cargo transport efficiency upon loss of interaction.motor regulation ͉ synaptic vesicle transport ͉ active zone protein ͉ axonal transport ͉ dynein

Nintedanib targets core features of SSc in Fra2-transgenic mice and ameliorates histological features of pulmonary arterial hypertension, destructive microangiopathy and pulmonary and dermal fibrosis. These data might have direct implications for the ongoing phase III clinical trial with nintedanib in SSc-associated interstitial lung disease.