The perceived blur of moving images is less than expected given the sluggish temporal response of the visual system. This suggests that a motion deblurring mechanism may exist to preserve the positional acuity and sharpness of moving images. Furthermore, when sequences of blurred stills are presented, observers report that the moving image is in sharp focus raising the possibility that there is...

We describe a primal-dual framework for the design and analysis of online convex optimization algorithms for drifting regret. Existing literature shows (nearly) optimal drifting regret bounds only for the l2 and the l1-norms. Our work provides a connection between these algorithms and the Online Mirror Descent (OMD) updates; one key insight that results from our work is that in order for these ...

Traditional data mining classifiers are used for mining the static data, in which incremental learning assumed data streams come under stationary distribution where data concepts remain unchanged. The concept of data can be changed at any time in real world application this refers to change in the class definitions over time. Classifier ensembles are rapidly gaining popularity in data mining Co...

There are high possibilities in the near future that great earthquakes occur and large-scale tsunamis attack Japan. In fishing ports facing on the ocean where a large number of fishing boats are moored, these boats would be littered in the ports, or drifting boats would collide with back-yard-structures, if such a tsunami attacks, like an event actually happened at Hokkaido in 1993. This is one...

In this work, we demonstrate an on-chip microfluidic flow cytometry system based on a three-dimensional (3D) hydrodynamic focusing technique, microfluidic drifting. By inducing Dean flow in a curved microfluidic channel, microfluidic drifting can be used to hydrodynamically focus cells or particles in the vertical direction and enables the 3D hydrodynamic focusing in a single-layer planar micro...

Sea-surface geostrophic velocities for the Kuroshio region calculated from TOPEX/ POSEIDON altimetry data together with in situ oceanographic data are compared with surface velocities derived from drifting buoy trajectories. The geostrophic velocities agree well with the observed velocities, suggesting that the Kuroshio surface layer is essentially in geostrophic balance, within measurement err...

Recently, on-line adaptation of binary classifiers for tracking have been investigated. On-line learning allows for simple classifiers since only the current view of the object from its surrounding background needs to be discriminiated. However, on-line adaption faces one key problem: Each update of the tracker may introduce an error which, finally, can lead to tracking failure (drifting). The ...

When humans view a moving object, the spatial lag in perception expected from neural delays may be partially corrected by motion mechanisms biasing perceived position. The drifting-Gabor illusion seems to support this view: the perceived location of a static envelope filled with a moving pattern is shifted in the direction of motion. To test whether this shifting mechanism also extrapolates the...

We present a new analysis of the problem of learning with drifting distributions in the batch setting using the notion of discrepancy. We prove learning bounds based on the Rademacher complexity of the hypothesis set and the discrepancy of distributions both for a drifting PAC scenario and a tracking scenario. Our bounds are always tighter and in some cases substantially improve upon previous o...

We investigated temporal aspects of the cortical mechanisms supporting visual contour integration by measuring observers' efficiency at detecting fragmented contours, composed of Gabor micropatterns, embedded in a field of distractor elements. Gabors consisted of a static Gaussian enveloping a sinusoidal carrier which was temporally modulated by motion or counter-phase flicker. The elements for...