作者: 易達(dá)到 時間: 2025-3-21 23:01
Visual Saliency Detection for RGB-D Images with Generative Modelmodel. The depth feature map is extracted based on superpixel contrast computation with spatial priors. We model the depth saliency map by approximating the density of depth-based contrast features using a Gaussian distribution. Similar to the depth saliency computation, the colour saliency map is c作者: 傳染 時間: 2025-3-22 01:05 作者: Lumbar-Spine 時間: 2025-3-22 06:49 作者: 聲明 時間: 2025-3-22 12:07
Generalized Fusion Moves for Continuous Label Optimizationpixel lattices and seek to assign discrete or continuous values (or both) to each pixel such that a combined data term and a spatial smoothness prior are minimized. In this work we propose to minimize difficult energies using repeated generalized fusion moves. In contrast to standard fusion moves, t作者: 不給啤 時間: 2025-3-22 16:31 作者: 不給啤 時間: 2025-3-22 18:45
phi-LSTM: A Phrase-Based Hierarchical LSTM Model for Image Captioningbe their attributes, and recognize their relationships/interactions. In this paper, we propose a phrase-based hierarchical Long Short-Term Memory (phi-LSTM) model to generate image description. The proposed model encodes sentence as a sequence of combination of phrases and words, instead of a sequen作者: Nutrient 時間: 2025-3-22 22:56 作者: Demulcent 時間: 2025-3-23 04:21 作者: GLIB 時間: 2025-3-23 06:42
Using Gaussian Processes to Improve Zero-Shot Learning with Relative Attributesimage is expressed in terms of attributes that are relatively specified between different class pairs. However, for zero-shot learning the authors had assumed a simple Gaussian Mixture Model (GMM) that used the GMM based clustering to obtain the label for an unknown target test example. In this pape作者: anthesis 時間: 2025-3-23 10:31
MARVEL: A Large-Scale Image Dataset for Maritime Vesselss, such as cars, bird species, and aircrafts, have been increasing. The collection of large datasets has helped vision based classification approaches and led to significant improvements in performances of the state-of-the-art methods. Visual classification of maritime vessels is another important t作者: 僵硬 時間: 2025-3-23 13:56 作者: Allowance 時間: 2025-3-23 20:11 作者: EXCEL 時間: 2025-3-24 01:10
R-CNN for Small Object Detectionring a small part of an image is largely ignored. As a result, the state-of-the-art object detection algorithm renders unsatisfactory performance as applied to detect small objects in images. In this paper, we dedicate an effort to bridge the gap. We first compose a benchmark dataset tailored for th作者: 吼叫 時間: 2025-3-24 03:32 作者: extract 時間: 2025-3-24 10:01
Object-Centric Representation Learning from Unlabeled Videosf (1) intensive manual annotations and (2) an inherent restriction in the scope of data relevant for learning. In this work, we explore unsupervised feature learning from unlabeled video. We introduce a novel . approach to temporal coherence that encourages similar representations to be learned for 作者: Fortuitous 時間: 2025-3-24 11:54 作者: 全國性 時間: 2025-3-24 15:44
Erratum to: Gentile da Fabriano,he fusion step optimizes over binary and continuous sets of variables representing label ranges. Further, each fusion step can optimize over additional continuous unknowns. We demonstrate the general method on a variational-inspired stereo approach, and optionally optimize over radiometric changes between the images as well.作者: 價值在貶值 時間: 2025-3-24 19:38
Conference proceedings 2017ce and Gestures; Image Alignment; Computational Photography and Image Processing; Language and Video; 3D Computer Vision; Image Attributes, Language, and Recognition; Video Understanding; and 3D Vision..作者: 滋養(yǎng) 時間: 2025-3-25 02:40
0302-9743 tions; Faces; Computational Photography; Face and Gestures; Image Alignment; Computational Photography and Image Processing; Language and Video; 3D Computer Vision; Image Attributes, Language, and Recognition; Video Understanding; and 3D Vision..978-3-319-54192-1978-3-319-54193-8Series ISSN 0302-9743 Series E-ISSN 1611-3349 作者: 要素 時間: 2025-3-25 06:11
https://doi.org/10.1007/978-94-015-2792-7cond stage, we formulate an optimization framework that enforces several constraints such as layout contour straightness, surface smoothness and geometric constraints for layout detail refinement. Our proposed system offers the state-of-the-art performance on two commonly used benchmark datasets.作者: 乳白光 時間: 2025-3-25 07:41
https://doi.org/10.1007/978-94-015-2798-9 features and the LSTM to learn the word sequence in a sentence, the proposed model has shown better or competitive results in comparison to the state-of-the-art models on Flickr8k and Flickr30k datasets.作者: 事物的方面 時間: 2025-3-25 14:43
https://doi.org/10.1007/978-94-015-0933-6and show that such a principled approach yields improved performance and a better understanding in terms of probabilistic estimates. The method is evaluated on standard Pubfig and Shoes with Attributes benchmarks.作者: 使迷醉 時間: 2025-3-25 18:33
Erratum to: Carlo and Vittorio Crivelli,culate these efficiently for mAP following NMS, enabling to train a detector based on Fast R-CNN?[.] directly for mAP. This model achieves equivalent performance to the standard Fast R-CNN on the PASCAL VOC 2007 and 2012 datasets, while being conceptually more appealing as the very same model and loss are used at both training and test time.作者: 搖曳 時間: 2025-3-25 22:00 作者: VEN 時間: 2025-3-26 01:02
https://doi.org/10.1007/978-94-015-2794-1racking required) while still being able to extract object-level regions from which to learn invariances. Furthermore, as we show in results on several standard datasets, our method typically achieves substantial accuracy gains over competing unsupervised methods for image classification and retrieval tasks.作者: Folklore 時間: 2025-3-26 06:36
https://doi.org/10.1007/978-1-349-10606-6datasets, where we obtain competitive or state-of-the-art results: on Stanford-40 Actions, we set a new state-of the art of 81.74%. On FGVC-Aircraft and the Stanford Dogs dataset, we show consistent improvements over baselines, some of which include significantly more supervision.作者: 凹處 時間: 2025-3-26 08:40
A Coarse-to-Fine Indoor Layout Estimation (CFILE) Methodcond stage, we formulate an optimization framework that enforces several constraints such as layout contour straightness, surface smoothness and geometric constraints for layout detail refinement. Our proposed system offers the state-of-the-art performance on two commonly used benchmark datasets.作者: atrophy 時間: 2025-3-26 13:52 作者: 光明正大 時間: 2025-3-26 17:43
Using Gaussian Processes to Improve Zero-Shot Learning with Relative Attributesand show that such a principled approach yields improved performance and a better understanding in terms of probabilistic estimates. The method is evaluated on standard Pubfig and Shoes with Attributes benchmarks.作者: eardrum 時間: 2025-3-27 00:54 作者: textile 時間: 2025-3-27 02:33
R-CNN for Small Object Detection for studying various design choices. Experiment results show that the augmented R-CNN algorithm improves the mean average precision by 29.8% over the original R-CNN algorithm on detecting small objects.作者: 外來 時間: 2025-3-27 05:49 作者: 獨(dú)裁政府 時間: 2025-3-27 11:12
Visual Concept Recognition and Localization via Iterative Introspectiondatasets, where we obtain competitive or state-of-the-art results: on Stanford-40 Actions, we set a new state-of the art of 81.74%. On FGVC-Aircraft and the Stanford Dogs dataset, we show consistent improvements over baselines, some of which include significantly more supervision.作者: Ethics 時間: 2025-3-27 15:07 作者: FLAX 時間: 2025-3-27 20:43
Generalized Fusion Moves for Continuous Label Optimizationhe fusion step optimizes over binary and continuous sets of variables representing label ranges. Further, each fusion step can optimize over additional continuous unknowns. We demonstrate the general method on a variational-inspired stereo approach, and optionally optimize over radiometric changes between the images as well.作者: Facilities 時間: 2025-3-27 22:55 作者: VEN 時間: 2025-3-28 05:35 作者: 木質(zhì) 時間: 2025-3-28 07:57 作者: 江湖騙子 時間: 2025-3-28 12:30
Computer Vision –ACCV 2016978-3-319-54193-8Series ISSN 0302-9743 Series E-ISSN 1611-3349 作者: cylinder 時間: 2025-3-28 15:38
https://doi.org/10.1007/978-94-015-2813-9 frame-to-frame correspondences, the tracking traditionally relies on the iterative closest point technique which does not scale well with the number of points. In this paper, we build on top of more recent and efficient density distribution alignment methods, and notably push the idea towards a hig作者: 繁忙 時間: 2025-3-28 20:20
https://doi.org/10.1007/978-94-015-2813-9model. The depth feature map is extracted based on superpixel contrast computation with spatial priors. We model the depth saliency map by approximating the density of depth-based contrast features using a Gaussian distribution. Similar to the depth saliency computation, the colour saliency map is c作者: Observe 時間: 2025-3-29 02:39
https://doi.org/10.1007/978-94-015-2792-7blem largely rely on hand-crafted features and vanishing lines, and they often fail in highly cluttered indoor scenes. The proposed coarse-to-fine indoor layout estimation (CFILE) method consists of two stages: (1) coarse layout estimation; and (2) fine layout localization. In the first stage, we ad作者: 罵人有污點(diǎn) 時間: 2025-3-29 05:38 作者: faddish 時間: 2025-3-29 07:36 作者: 笨拙的我 時間: 2025-3-29 14:01
Raimond Van Marle,Charlotte Van Marlethe contextual relationship in the image, such as the kind of object, relationship between two objects, or the action. In this paper, we turn our attention to more subjective components of descriptions that contain rich expressions to modify objects – namely attribute expressions. We start by collec作者: Anecdote 時間: 2025-3-29 17:00
https://doi.org/10.1007/978-94-015-2798-9be their attributes, and recognize their relationships/interactions. In this paper, we propose a phrase-based hierarchical Long Short-Term Memory (phi-LSTM) model to generate image description. The proposed model encodes sentence as a sequence of combination of phrases and words, instead of a sequen作者: 知道 時間: 2025-3-29 20:15
https://doi.org/10.1007/978-94-015-2798-9e between images and to be able to compare the strength of each property between images, relative attributes were introduced. However, since their introduction, hand-crafted and engineered features were used to learn increasingly complex models for the problem of relative attributes. This limits the作者: 自戀 時間: 2025-3-30 02:48 作者: arsenal 時間: 2025-3-30 05:21 作者: 事先無準(zhǔn)備 時間: 2025-3-30 11:45
https://doi.org/10.1007/978-94-015-2796-5s, such as cars, bird species, and aircrafts, have been increasing. The collection of large datasets has helped vision based classification approaches and led to significant improvements in performances of the state-of-the-art methods. Visual classification of maritime vessels is another important t作者: Herpetologist 時間: 2025-3-30 12:37 作者: 不舒服 時間: 2025-3-30 18:01
Erratum to: Carlo and Vittorio Crivelli,end fashion that includes non-maximum suppresion (NMS) at training time. This contrasts with the traditional approach of training a CNN for a window classification loss, then applying NMS only at test time, when mAP is used as the evaluation metric in place of classification accuracy. However, mAP f作者: prediabetes 時間: 2025-3-31 00:25
Erratum to: Carlo and Vittorio Crivelli,ring a small part of an image is largely ignored. As a result, the state-of-the-art object detection algorithm renders unsatisfactory performance as applied to detect small objects in images. In this paper, we dedicate an effort to bridge the gap. We first compose a benchmark dataset tailored for th作者: 畸形 時間: 2025-3-31 02:40
https://doi.org/10.1007/978-94-015-2794-1ssification. More specifically, a triplet is created among “three” whole templates or subtemplates of images to incorporate the (sub)template structure into metric learning. To further account for intra-class variations of images, we introduce a factorization technique to integrate image-specific co作者: MAOIS 時間: 2025-3-31 06:24 作者: RAGE 時間: 2025-3-31 12:55 作者: 專心 時間: 2025-3-31 13:51
https://doi.org/10.1007/978-3-319-54193-83D vision; clustering; computer vision; image processing; neural networks; action recognition; computation作者: Pseudoephedrine 時間: 2025-3-31 17:38 作者: Fallibility 時間: 2025-3-31 23:13 作者: Incorruptible 時間: 2025-4-1 04:46
Lecture Notes in Computer Sciencehttp://image.papertrans.cn/c/image/234115.jpg作者: artifice 時間: 2025-4-1 09:28
Divide and Conquer: Efficient Density-Based Tracking of 3D Sensors in Manhattan Worlds-shift paradigm to manifold-constrained multiple-mode tracking. Dedicated projections subsequently enable the estimation of the translation through three simple 1D density alignment steps that can be executed in parallel. An extensive evaluation on both simulated and publicly available real datasets
