The Experts below are selected from a list of 5739 Experts worldwide ranked by ideXlab platform
Alykhan Tejani - One of the best experts on this subject based on the ideXlab platform.
-
Latent Regression Forest: Structured Estimation of 3D Hand Poses
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017Co-Authors: Danhang Tang, Hyung Jin Chang, Alykhan TejaniAbstract:In this paper we present the latent regression forest (LRF), a novel framework for real-time, 3D hand pose estimation from a single depth image. Prior discriminative methods often fall into two categories: holistic and patch-based. Holistic methods are efficient but less flexible due to their nearest neighbour nature. Patch-based methods can generalise to unseen samples by consider local appearance only. However, they are complex because each pixel need to be classified or regressed during testing. In contrast to these two baselines, our method can be considered as a Structured coarse-to-fine Search, starting from the centre of mass of a point cloud until locating all the skeletal joints. The Searching process is guided by a learnt latent tree model which reflects the hierarchical topology of the hand. Our main contributions can be summarised as follows: (i) Learning the topology of the hand in an unsupervised, data-driven manner. (ii) A new forest-based, discriminative framework for Structured Search in images, as well as an error regression step to avoid error accumulation. (iii) A new multi-view hand pose dataset containing 180 K annotated images from 10 different subjects. Our experiments on two datasets show that the LRF outperforms baselines and prior arts in both accuracy and efficiency.
-
Latent Regression Forest: Structured Estimation of 3D Articulated Hand Posture
2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014Co-Authors: Danhang Tang, Hyung Jin Chang, Alykhan TejaniAbstract:In this paper we present the Latent Regression Forest (LRF), a novel framework for real-time, 3D hand pose estimation from a single depth image. In contrast to prior forest-based methods, which take dense pixels as input, classify them independently and then estimate joint positions afterwards, our method can be considered as a Structured coarse-to-fine Search, starting from the centre of mass of a point cloud until locating all the skeletal joints. The Searching process is guided by a learnt Latent Tree Model which reflects the hierarchical topology of the hand. Our main contributions can be summarised as follows: (i) Learning the topology of the hand in an unsupervised, data-driven manner. (ii) A new forest-based, discriminative framework for Structured Search in images, as well as an error regression step to avoid error accumulation. (iii) A new multi-view hand pose dataset containing 180K annotated images from 10 different subjects. Our experiments show that the LRF out-performs state-of-the-art methods in both accuracy and efficiency.
Danhang Tang - One of the best experts on this subject based on the ideXlab platform.
-
Latent Regression Forest: Structured Estimation of 3D Hand Poses
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017Co-Authors: Danhang Tang, Hyung Jin Chang, Alykhan TejaniAbstract:In this paper we present the latent regression forest (LRF), a novel framework for real-time, 3D hand pose estimation from a single depth image. Prior discriminative methods often fall into two categories: holistic and patch-based. Holistic methods are efficient but less flexible due to their nearest neighbour nature. Patch-based methods can generalise to unseen samples by consider local appearance only. However, they are complex because each pixel need to be classified or regressed during testing. In contrast to these two baselines, our method can be considered as a Structured coarse-to-fine Search, starting from the centre of mass of a point cloud until locating all the skeletal joints. The Searching process is guided by a learnt latent tree model which reflects the hierarchical topology of the hand. Our main contributions can be summarised as follows: (i) Learning the topology of the hand in an unsupervised, data-driven manner. (ii) A new forest-based, discriminative framework for Structured Search in images, as well as an error regression step to avoid error accumulation. (iii) A new multi-view hand pose dataset containing 180 K annotated images from 10 different subjects. Our experiments on two datasets show that the LRF outperforms baselines and prior arts in both accuracy and efficiency.
-
Latent Regression Forest: Structured Estimation of 3D Articulated Hand Posture
2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014Co-Authors: Danhang Tang, Hyung Jin Chang, Alykhan TejaniAbstract:In this paper we present the Latent Regression Forest (LRF), a novel framework for real-time, 3D hand pose estimation from a single depth image. In contrast to prior forest-based methods, which take dense pixels as input, classify them independently and then estimate joint positions afterwards, our method can be considered as a Structured coarse-to-fine Search, starting from the centre of mass of a point cloud until locating all the skeletal joints. The Searching process is guided by a learnt Latent Tree Model which reflects the hierarchical topology of the hand. Our main contributions can be summarised as follows: (i) Learning the topology of the hand in an unsupervised, data-driven manner. (ii) A new forest-based, discriminative framework for Structured Search in images, as well as an error regression step to avoid error accumulation. (iii) A new multi-view hand pose dataset containing 180K annotated images from 10 different subjects. Our experiments show that the LRF out-performs state-of-the-art methods in both accuracy and efficiency.
Hyung Jin Chang - One of the best experts on this subject based on the ideXlab platform.
-
Latent Regression Forest: Structured Estimation of 3D Hand Poses
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017Co-Authors: Danhang Tang, Hyung Jin Chang, Alykhan TejaniAbstract:In this paper we present the latent regression forest (LRF), a novel framework for real-time, 3D hand pose estimation from a single depth image. Prior discriminative methods often fall into two categories: holistic and patch-based. Holistic methods are efficient but less flexible due to their nearest neighbour nature. Patch-based methods can generalise to unseen samples by consider local appearance only. However, they are complex because each pixel need to be classified or regressed during testing. In contrast to these two baselines, our method can be considered as a Structured coarse-to-fine Search, starting from the centre of mass of a point cloud until locating all the skeletal joints. The Searching process is guided by a learnt latent tree model which reflects the hierarchical topology of the hand. Our main contributions can be summarised as follows: (i) Learning the topology of the hand in an unsupervised, data-driven manner. (ii) A new forest-based, discriminative framework for Structured Search in images, as well as an error regression step to avoid error accumulation. (iii) A new multi-view hand pose dataset containing 180 K annotated images from 10 different subjects. Our experiments on two datasets show that the LRF outperforms baselines and prior arts in both accuracy and efficiency.
-
Latent Regression Forest: Structured Estimation of 3D Articulated Hand Posture
2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014Co-Authors: Danhang Tang, Hyung Jin Chang, Alykhan TejaniAbstract:In this paper we present the Latent Regression Forest (LRF), a novel framework for real-time, 3D hand pose estimation from a single depth image. In contrast to prior forest-based methods, which take dense pixels as input, classify them independently and then estimate joint positions afterwards, our method can be considered as a Structured coarse-to-fine Search, starting from the centre of mass of a point cloud until locating all the skeletal joints. The Searching process is guided by a learnt Latent Tree Model which reflects the hierarchical topology of the hand. Our main contributions can be summarised as follows: (i) Learning the topology of the hand in an unsupervised, data-driven manner. (ii) A new forest-based, discriminative framework for Structured Search in images, as well as an error regression step to avoid error accumulation. (iii) A new multi-view hand pose dataset containing 180K annotated images from 10 different subjects. Our experiments show that the LRF out-performs state-of-the-art methods in both accuracy and efficiency.
Trond Aalberg - One of the best experts on this subject based on the ideXlab platform.
-
Branch Filtering of Tree-Structured Search Results
2019 ACM IEEE Joint Conference on Digital Libraries (JCDL), 2019Co-Authors: Trond AalbergAbstract:The use of categories to filter results is a major feature in modern Search systems. With the introduction of entity-centric information models for bibliographic data such as the IFLA Library Reference Model (LRM), the items returned for library Search typically have an internal tree-structure and filtering needs to be applied on branches. This contribution presents a solution that is demonstrated in a Search system for LRM-based data, but the approach is generic and can be adapted to the filtering of any tree-based Search results and combination logic.
-
JCDL - Branch Filtering of Tree-Structured Search Results
2019 ACM IEEE Joint Conference on Digital Libraries (JCDL), 2019Co-Authors: Trond AalbergAbstract:The use of categories to filter results is a major feature in modern Search systems. With the introduction of entity-centric information models for bibliographic data such as the IFLA Library Reference Model (LRM), the items returned for library Search typically have an internal tree-structure and filtering needs to be applied on branches. This contribution presents a solution that is demonstrated in a Search system for LRM-based data, but the approach is generic and can be adapted to the filtering of any tree-based Search results and combination logic.
Iman Rasekh - One of the best experts on this subject based on the ideXlab platform.
-
A new competitive intelligence-based strategy for web page Search
Proceedings of the 2015 IEEE 9th International Conference on Semantic Computing (IEEE ICSC 2015), 2015Co-Authors: Iman RasekhAbstract:Semantic Web is known as next generation of web it is known as a new collaborative movement toward Web3.0 that led by the World Wide Web Consortium (W3C). It aims at converting the current web of unStructured documents into a “web of data”. The proposed Searching strategy for SEO in Semantic Web is a graph Structured Search (GSS). Search Engine Optimization (SEO) is defined as a collection of techniques and practices that allow a site to get more traffic from Search engines and it is still one of the biggest challenge in Search engines of Semantic Webs In this paper, I proposed a new type of web page Search which is based on the competitive intelligence. It use link-based ranking evolutionary scheme to accommodate users' preferences. I implemented the prototype system and demonstrate the feasibility of the proposed web page Search scheme.
-
Dynamic Search Optimization for Semantic Webs Using Imperialistic Competitive Algorithm
2012 International Conference on Information Science and Applications, 2012Co-Authors: Iman RasekhAbstract:Web 3.0 is known as next generation of web technology after linear presentation of information (Web 1.0) and multi-linear presentation of information (Web 2.0) .Semantic Web is a new collaborative movement toward Web3.0 that led by the World Wide Web Consortium (W3C) .the Semantic Web aims at converting the current web of unStructured documents into a "web of data". Searching is a big challenge in semantic web. Its Searching strategy is graph Structured Search (GSS). It is quite different with the current structure of Web documents. GSS defines as a graph of relationships between anchor nodes but still there is no sufficient Search engine designed for it. In this paper we tried to use imperialistic Competitive Algorithm (ICA) to improve Semantic Web Searching. By using ICA, a Structured scheme, a pow.erful semantic Search engine could be designed.
-
Dynamic Search optimization for semantic webs using imperialistic competitive algorithm
2012 International Conference on Information Science and Applications ICISA 2012, 2012Co-Authors: Iman RasekhAbstract:Web 3.0 is known as next generation of web technology after linear presentation of information (Web 1.0) and multi-linear presentation of information (Web 2.0) .Semantic Web is a new collaborative movement toward Web3.0 that led by the World Wide Web Consortium (W3C) .the Semantic Web aims at converting the current web of unStructured documents into a "web of data". Searching is a big challenge in semantic web. Its Searching strategy is graph Structured Search (GSS). It is quite different with the current structure of Web documents. GSS defines as a graph of relationships between anchor nodes but still there is no sufficient Search engine designed for it. In this paper we tried to use imperialistic Competitive Algorithm (ICA) to improve Semantic Web Searching. By using ICA, a Structured scheme, a pow.erful semantic Search engine could be designed. View full abstract»