The proceedings contain 602 papers. The topics discussed include: going deeper with convolutions;propagated image filtering;web scale photo hash clustering on a single machine;supervised discrete hashing;what do 15,00...
ISBN:
(纸本)9781467369640
The proceedings contain 602 papers. The topics discussed include: going deeper with convolutions;propagated image filtering;web scale photo hash clustering on a single machine;supervised discrete hashing;what do 15,000 object categories tell us about classifying and localizing actions?;landmarks-based kernelized subspace alignment for unsupervised domain adaptation;blur kernel estimation using normalized color-line priors;a light transport model for mitigating multipath interference in time-of-flight sensors;traditional saliency reloaded: a good old model in new shape;automatic construction of robust spherical harmonic subspaces;leveraging stereo matching with learning-based confidence measures;saliency detection via cellular automata;and efficient sparse-to-dense optical flow estimation using a learned basis and layers.
The proceedings contain 539 papers. The topics discussed include: fast and accurate image matching with cascade hashing for 3D reconstruction;minimal solvers for relative pose with a single unknown radial distortion;s...
ISBN:
(纸本)9781479951178
The proceedings contain 539 papers. The topics discussed include: fast and accurate image matching with cascade hashing for 3D reconstruction;minimal solvers for relative pose with a single unknown radial distortion;spectral graph reduction for efficient image and streaming video segmentation;video motion segmentation using new adaptive manifold denoising model;event detection using multi-level relevance labels and multiple features;full-angle quaternions for robustly matching vectors of 3D rotations;semi-supervised spectral clustering for image set classification;learning mid-level filters for person re-identification;DeepReID: deep filter pairing neural network for person re-identification;NMF-KNN: image annotation using weighted multi-view non-negative matrix factorization;beyond comparing image pairs: setwise active learning for relative attributes;and histograms of pattern sets for image classification and object recognition.
The proceedings contain 979 papers. The topics discussed include: embodied question answering;learning by asking questions;finding tiny faces in the wild with generative adversarial network;paired CycleGAN: asymmetric...
ISBN:
(纸本)9781538664209
The proceedings contain 979 papers. The topics discussed include: embodied question answering;learning by asking questions;finding tiny faces in the wild with generative adversarial network;paired CycleGAN: asymmetric style transfer for applying and removing makeup;learning pose specific representations by predicting different views;weakly and semi supervised human body part parsing via pose-guided knowledge transfer;person transfer GAN to bridge domain gap for person re-identification;cross-modal deep variational hand pose estimation;disentangled person image generation;super-fan: integrated facial landmark localization and super-resolution of real-world low resolution faces in arbitrary poses with GANs;multistage adversarial losses for pose-based human image synthesis;and a certifiably globally optimal solution to the non-minimal relative pose problem.
The proceedings contain 781 papers. The topics discussed include: exclusivity-consistency regularized multi-view subspace clustering;borrowing treasures from the wealthy: deep transfer learning through selective joint...
The proceedings contain 781 papers. The topics discussed include: exclusivity-consistency regularized multi-view subspace clustering;borrowing treasures from the wealthy: deep transfer learning through selective joint fine-tuning;the more you know: using knowledge graphs for image classification;dynamic edge-conditioned filters in convolutional neural networks on graphs;convolutional neural network architecture for geometric matching;deep affordance-grounded sensorimotor object recognition;on compressing deep models by low rank and sparse decomposition;unsupervised pixel-level domain adaptation with generative adversarial networks;photo-realistic single image super-resolution using a generative adversarial network;a practical method for fully automatic intrinsic camera calibration using directionally encoded light;elastic shape-from-template with spatially sparse deforming forces;and distinguishing the indistinguishable: exploring structural ambiguities via geodesic context.
The proceedings contain 626 papers. The topics discussed include: learning to match aerial images with deep attentive architectures; learnt quasi-transitive similarity for retrieval from large collections of faces; un...
ISBN:
(纸本)9781467388511
The proceedings contain 626 papers. The topics discussed include: learning to match aerial images with deep attentive architectures; learnt quasi-transitive similarity for retrieval from large collections of faces; unsupervised learning from narrated instruction videos; action recognition in video using sparse coding and relative features; recognizing emotions from abstract paintings using non-linear matrix completion; rolling shutter absolute pose problem with known vertical direction; memory efficient max flow for multi-label submodular MRFs; recovering the missing link: predicting class-attribute associations for unsupervised zero-shot learning; neural module networks; multi-cue zero-shot learning with strong supervision; social LSTM: human trajectory prediction in crowded spaces; NetVLAD: CNN architecture for weakly supervised place recognition; 3D semantic parsing of large-scale indoor spaces; efficient indexing of billion-scale datasets of deep descriptors; analyzing classifiers: Fisher vectors and deep neural networks; and GIFT: a real-time and scalable 3d shape search engine.
The proceedings contain 640 papers. The topics discussed include: generation and comprehension of unambiguous object description;image question answering using convolutional neural network with dynamic parameter predi...
ISBN:
(纸本)9781467388504
The proceedings contain 640 papers. The topics discussed include: generation and comprehension of unambiguous object description;image question answering using convolutional neural network with dynamic parameter prediction;neural module networks;learning to assign orientations to feature points;affinity CNN: learning pixel-centric pairwise relations for figure/ground embedding;occlusion boundary detection via deep exploration of context;exploit bounding box annotations for multi-label object recognition;MCMC shape sampling for image segmentation with nonparametric shape priors;learning action maps of large environments via first-person vision;sample and filter: nonparametric scene parsing via efficient filtering;training region-based object detectors with online hard example mining;learning with side information through modality hallucination;HyperNet: towards accurate region proposal generation and joint object detection;macroscopic interferometry: rethinking depth estimation with frequency-domain time-of-flight;ASP vision: optically computing the first layer of convolutional neural networks using angle sensitive pixels;hierarchical recurrent neural encoder for video representation with application to captioning;and from keyframes to key objects: video summarization by representative object proposal selection.
暂无评论