A biologically inspired appearance model for robust visual tracking

In this work, we propose a biologically inspired appearance model for robust visual tracking. Motivated in part by the success of the hierarchical organization of the primary visual cortex (area V1), we establish an architecture consisting of five layers: whitening, rectification, normalization, coding and polling. The first three layers stem from the models developed for object recognition. In this paper, our attention focuses on the coding and pooling layers. In particular, we use a discriminative sparse coding method in the coding layer along with spatial pyramid representation in the pooling layer, which makes it easier to distinguish the target to be tracked from its background in the presence of appearance variations. An extensive experimental study shows that the proposed method has higher tracking accuracy than several state-of-the-art trackers.

Feature recognition & tool path generation for 5 axis step-nc machining of free form / irregular contoured surfaces

This research paper presents a five step algorithm to generate tool paths for machining Free form / Irregular Contoured Surface(s) (FICS) by adopting STEP-NC (AP-238) format. In the first step, a parametrized CAD model with FICS is created or imported in UG-NX6.0 CAD package. The second step recognizes the features and calculates a Closeness Index (CI) by comparing them with the B-Splines / Bezier surfaces. The third step utilizes the CI and extracts the necessary data to formulate the blending functions for identified features. In the fourth step Z-level 5 axis tool paths are generated by adopting flat and ball end mill cutters. Finally, in the fifth step, tool paths are integrated with STEP-NC format and validated. All these steps are discussed and explained through a validated industrial component.

On covariate factor detection and removal for robust gait recognition

We propose a novel bolt-on module capable of boosting the robustness of various single compact 2D gait representations. Gait recognition is negatively influenced by covariate factors including clothing and time which alter the natural gait appearance and motion. Contrary to traditional gait recognition, our bolt-on module remedies this by a dedicated covariate factor detection and removal procedure which we quantitatively and qualitatively evaluate. The fundamental concept of the bolt-on module is founded on exploiting the pixel-wise composition of covariate factors. Results demonstrate how our bolt-on module is a powerful component leading to significant improvements across gait representations and datasets yielding state-of-the-art results.

Discovery of a New Retrograde Trans-Neptunian Object: Hint of a Common Orbital Plane for Low Semimajor Axis, High-inclination TNOs and Centaurs

Although the majority of Centaurs are thought to have originated in the scattered disk, with the high-inclination members coming from the Oort cloud, the origin of the high-inclination component of trans-Neptunian objects (TNOs) remains uncertain. We report the discovery of a retrograde TNO, which we nickname “Niku,” detected by the Pan-STARRS 1 Outer Solar System Survey. Our numerical integrations show that the orbital dynamics of Niku are very similar to that of 2008 KV42 (Drac), with a half-life of ˜500 Myr. Comparing similar high-inclination TNOs and Centaurs (q > 10 au, a <100 au, and i > 60°), we find that these objects exhibit a surprising clustering of ascending node, and occupy a common orbital plane. This orbital configuration has high statistical significance: 3.8-σ. An unknown mechanism is required to explain the observed clustering. This discovery may provide a pathway to investigating a possible reservoir of high-inclination objects.