{"title":"Zero-Shot Object Recognition Using Semantic Label Vectors","authors":"Shujon Naha, Yang Wang","doi":"10.1109/CRV.2015.21","DOIUrl":"https://doi.org/10.1109/CRV.2015.21","url":null,"abstract":"We consider the problem of zero-shot recognition of object categories from images. Given a set of object categories (called \"known classes\") with training images, our goal is to learn a system to recognize another non-overlapping set of object categories (called \"unknown classes\") for which there are no training images. Our proposed approach exploits the recent work in natural language processing which has produced vector representations of words. Using the vector representations of object classes, we develop a method for transferring the appearance models from known object classes to unknown object classes. Our experimental results on three benchmark datasets show that our proposed method outperforms other competing approaches.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114173158","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Lee Clement, Valentin Peretroukhin, Jacob Lambert, Jonathan Kelly
{"title":"The Battle for Filter Supremacy: A Comparative Study of the Multi-State Constraint Kalman Filter and the Sliding Window Filter","authors":"Lee Clement, Valentin Peretroukhin, Jacob Lambert, Jonathan Kelly","doi":"10.1109/CRV.2015.11","DOIUrl":"https://doi.org/10.1109/CRV.2015.11","url":null,"abstract":"Accurate and consistent ego motion estimation is a critical component of autonomous navigation. For this task, the combination of visual and inertial sensors is an inexpensive, compact, and complementary hardware suite that can be used on many types of vehicles. In this work, we compare two modern approaches to ego motion estimation: the Multi-State Constraint Kalman Filter (MSCKF) and the Sliding Window Filter (SWF). Both filters use an Inertial Measurement Unit (IMU) to estimate the motion of a vehicle and then correct this estimate with observations of salient features from a monocular camera. While the SWF estimates feature positions as part of the filter state itself, the MSCKF optimizes feature positions in a separate procedure without including them in the filter state. We present experimental characterizations and comparisons of the MSCKF and SWF on data from a moving hand-held sensor rig, as well as several traverses from the KITTI dataset. In particular, we compare the accuracy and consistency of the two filters, and analyze the effect of feature track length and feature density on the performance of each filter. In general, our results show the SWF to be more accurate and less sensitive to tuning parameters than the MSCKF. However, the MSCKF is computationally cheaper, has good consistency properties, and improves in accuracy as more features are tracked.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"130 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122643572","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Latent SVM for Object Localization in Weakly Labeled Videos","authors":"Mrigank Rochan, Yang Wang","doi":"10.1109/CRV.2015.33","DOIUrl":"https://doi.org/10.1109/CRV.2015.33","url":null,"abstract":"We consider the problem of object localization in Internet videos. An Internet video (e.g. YouTube videos) is often associated with a semantic label (also known as a tag) describing the main object present within it. However, the tag does not provide any spatial or temporal information about the main object in the video. Such videos are weakly labelled. Given weakly labelled video with video-level object class tags, our goal is to learn a model that can be used to localize the objects in other videos with such tags. We define a latent SVM based learning framework to tackle this problem. We demonstrate the effectiveness of our method on a dataset composed of videos collected from YouTube.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122840608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Mobile 3D Gaze Tracking Calibration","authors":"Christian Scheel, O. Staadt","doi":"10.1109/CRV.2015.30","DOIUrl":"https://doi.org/10.1109/CRV.2015.30","url":null,"abstract":"We present a new calibration method to combine a mobile eye tracker with an external tracking system to obtain a 3D gaze vector. Our method captures calibration points of varying distances, pupil positions and head positions/orientations. With these data we can determine the eye position relative to the user's head position without separate manual eye-position measurements. For this approach, it is not necessary to know the orientation of the eye coordinate system in advance. In addition to the calibration of the external tracking system calibration, we can calibrate the head-tracked eye tracker in a one-step process, requiring the user to look at the calibration points. No extra calibration of the eye tracker is necessary, if the raw pupil position in the eye-camera is available from the eye tracker. The calibrated system allows us to estimate the 3D gaze vector for a user who can move freely within the range of the external tracking system. Our evaluation shows that the average accuracy of the visual angle is better than one degree in a self evaluation and approximately two degrees under unrestrained head movement.","PeriodicalId":257595,"journal":{"name":"2015 12th Conference on Computer and Robot Vision","volume":"151 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127288554","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}