| Total: 158
Most counting questions in visual question answering (VQA) datasets are simple and require no more than object detection. Here, we study algorithms for complex counting questions that involve relationships between objects, attribute identification, reasoning, and more. To do this, we created TallyQA, the world’s largest dataset for open-ended counting. We propose a new algorithm for counting that uses relation networks with region proposals. Our method lets relation networks be efficiently used with high-resolution imagery. It yields stateof-the-art results compared to baseline and recent systems on both TallyQA and the HowMany-QA benchmark.
This paper presents Densely Supervised Grasp Detector (DSGD), a deep learning framework which combines CNN structures with layer-wise feature fusion and produces grasps and their confidence scores at different levels of the image hierarchy (i.e., global-, region-, and pixel-levels). Specifically, at the global-level, DSGD uses the entire image information to predict a grasp. At the region-level, DSGD uses a region proposal network to identify salient regions in the image and uses a grasp prediction network to generate segmentations and their corresponding grasp poses of the salient regions. At the pixel-level, DSGD uses a fully convolutional network and predicts a grasp and its confidence at every pixel. During inference, DSGD selects the most confident grasp as the output. This selection from hierarchically generated grasp candidates overcomes limitations of the individual models. DSGD outperforms state-of-the-art methods on the Cornell grasp dataset in terms of grasp accuracy. Evaluation on a multi-object dataset and real-world robotic grasping experiments show that DSGD produces highly stable grasps on a set of unseen objects in new environments. It achieves 97% grasp detection accuracy and 90% robotic grasping success rate with real-time inference speed.
Region-based object detection infers object regions for one or more categories in an image. Due to the recent advances in deep learning and region proposal methods, object detectors based on convolutional neural networks (CNNs) have been flourishing and provided the promising detection results. However, the detection accuracy is degraded often because of the low discriminability of object CNN features caused by occlusions and inaccurate region proposals. In this paper, we therefore propose a region decomposition and assembly detector (R-DAD) for more accurate object detection. In the proposed R-DAD, we first decompose an object region into multiple small regions. To capture an entire appearance and part details of the object jointly, we extract CNN features within the whole object region and decomposed regions. We then learn the semantic relations between the object and its parts by combining the multi-region features stage by stage with region assembly blocks, and use the combined and high-level semantic features for the object classification and localization. In addition, for more accurate region proposals, we propose a multi-scale proposal layer that can generate object proposals of various scales. We integrate the R-DAD into several feature extractors, and prove the distinct performance improvement on PASCAL07/12 and MSCOCO18 compared to the recent convolutional detectors.
Multimodal representation learning is gaining more and more interest within the deep learning community. While bilinear models provide an interesting framework to find subtle combination of modalities, their number of parameters grows quadratically with the input dimensions, making their practical implementation within classical deep learning pipelines challenging. In this paper, we introduce BLOCK, a new multimodal fusion based on the block-superdiagonal tensor decomposition. It leverages the notion of block-term ranks, which generalizes both concepts of rank and mode ranks for tensors, already used for multimodal fusion. It allows to define new ways for optimizing the tradeoff between the expressiveness and complexity of the fusion model, and is able to represent very fine interactions between modalities while maintaining powerful mono-modal representations. We demonstrate the practical interest of our fusion model by using BLOCK for two challenging tasks: Visual Question Answering (VQA) and Visual Relationship Detection (VRD), where we design end-to-end learnable architectures for representing relevant interactions between modalities. Through extensive experiments, we show that BLOCK compares favorably with respect to state-of-the-art multimodal fusion models for both VQA and VRD tasks. Our code is available at https://github.com/Cadene/block.bootstrap.pytorch.
Inferring the interactions between objects, a.k.a visual relationship detection, is a crucial point for vision understanding, which captures more definite concepts than object detection. Most previous work that treats the interaction between a pair of objects as a one way fail to exploit the mutual relation between objects, which is essential to modern visual application. In this work, we propose a mutual relation net, dubbed MR-Net, to explore the mutual relation between paired objects for visual relationship detection. Specifically, we construct a mutual relation space to model the mutual interaction of paired objects, and employ linear constraint to optimize the mutual interaction, which is called mutual relation learning. Our mutual relation learning does not introduce any parameters, and can adapt to improve the performance of other methods. In addition, we devise a semantic ranking loss to discriminatively penalize predicates with semantic similarity, which is ignored by traditional loss function (e.g., cross entropy with softmax). Then, our MR-Net optimizes the mutual relation learning together with semantic ranking loss with a siamese network. The experimental results on two commonly used datasets (VG and VRD) demonstrate the superior performance of the proposed approach.
Predicting action class from partially observed videos, which is known as action prediction, is an important task in computer vision field with many applications. The challenge for action prediction mainly lies in the lack of discriminative action information for the partially observed videos. To tackle this challenge, in this work, we propose to transfer action knowledge learned from fully observed videos for improving the prediction of partially observed videos. Specifically, we develop a two-stage learning framework for action knowledge transfer. At the first stage, we learn feature embeddings and discriminative action classifier from full videos. The knowledge in the learned embeddings and classifier is then transferred to the partial videos at the second stage. Our experiments on the UCF-101 and HMDB-51 datasets show that the proposed action knowledge transfer method can significantly improve the performance of action prediction, especially for the actions with small observation ratios (e.g., 10%). We also experimentally illustrate that our method outperforms all the state-of-the-art action prediction systems.
As a unique biometric feature that can be recognized at a distance, gait has broad applications in crime prevention, forensic identification and social security. To portray a gait, existing gait recognition methods utilize either a gait template, where temporal information is hard to preserve, or a gait sequence, which must keep unnecessary sequential constraints and thus loses the flexibility of gait recognition. In this paper we present a novel perspective, where a gait is regarded as a set consisting of independent frames. We propose a new network named GaitSet to learn identity information from the set. Based on the set perspective, our method is immune to permutation of frames, and can naturally integrate frames from different videos which have been filmed under different scenarios, such as diverse viewing angles, different clothes/carrying conditions. Experiments show that under normal walking conditions, our single-model method achieves an average rank-1 accuracy of 95.0% on the CASIA-B gait dataset and an 87.1% accuracy on the OU-MVLP gait dataset. These results represent new state-of-the-art recognition accuracy. On various complex scenarios, our model exhibits a significant level of robustness. It achieves accuracies of 87.2% and 70.4% on CASIA-B under bag-carrying and coat-wearing walking conditions, respectively. These outperform the existing best methods by a large margin. The method presented can also achieve a satisfactory accuracy with a small number of frames in a test sample, e.g., 82.5% on CASIA-B with only 7 frames. The source code has been released at https://github.com/AbnerHqC/GaitSet.
Deep metric learning has been widely applied in many computer vision tasks, and recently, it is more attractive in zeroshot image retrieval and clustering (ZSRC) where a good embedding is requested such that the unseen classes can be distinguished well. Most existing works deem this ’good’ embedding just to be the discriminative one and thus race to devise powerful metric objectives or hard-sample mining strategies for leaning discriminative embedding. However, in this paper, we first emphasize that the generalization ability is a core ingredient of this ’good’ embedding as well and largely affects the metric performance in zero-shot settings as a matter of fact. Then, we propose the Energy Confused Adversarial Metric Learning (ECAML) framework to explicitly optimize a robust metric. It is mainly achieved by introducing an interesting Energy Confusion regularization term, which daringly breaks away from the traditional metric learning idea of discriminative objective devising, and seeks to ’confuse’ the learned model so as to encourage its generalization ability by reducing overfitting on the seen classes. We train this confusion term together with the conventional metric objective in an adversarial manner. Although it seems weird to ’confuse’ the network, we show that our ECAML indeed serves as an efficient regularization technique for metric learning and is applicable to various conventional metric methods. This paper empirically and experimentally demonstrates the importance of learning embedding with good generalization, achieving state-of-theart performances on the popular CUB, CARS, Stanford Online Products and In-Shop datasets for ZSRC tasks. Code available at http://www.bhchen.cn/.
In this paper, we propose a novel conditional-generativeadversarial-nets-based image captioning framework as an extension of traditional reinforcement-learning (RL)-based encoder-decoder architecture. To deal with the inconsistent evaluation problem among different objective language metrics, we are motivated to design some “discriminator” networks to automatically and progressively determine whether generated caption is human described or machine generated. Two kinds of discriminator architectures (CNN and RNNbased structures) are introduced since each has its own advantages. The proposed algorithm is generic so that it can enhance any existing RL-based image captioning framework and we show that the conventional RL training method is just a special case of our approach. Empirically, we show consistent improvements over all language evaluation metrics for different state-of-the-art image captioning models. In addition, the well-trained discriminators can also be viewed as objective image captioning evaluators.
Most of the existing works on image description focus on generating expressive descriptions. The only few works that are dedicated to generating stylish (e.g., romantic, lyric, etc.) descriptions suffer from limited style variation and content digression. To address these limitations, we propose a controllable stylish image description generation model. It can learn to generate stylish image descriptions that are more related to image content and can be trained with the arbitrary monolingual corpus without collecting new paired image and stylish descriptions. Moreover, it enables users to generate various stylish descriptions by plugging in style-specific parameters to include new styles into the existing model. We achieve this capability via a novel layer normalization layer design, which we will refer to as the Domain Layer Norm (DLN). Extensive experimental validation and user study on various stylish image description generation tasks are conducted to show the competitive advantages of the proposed model.
This paper presents a “learning to learn” approach to figureground image segmentation. By exploring webly-abundant images of specific visual effects, our method can effectively learn the visual-effect internal representations in an unsupervised manner and uses this knowledge to differentiate the figure from the ground in an image. Specifically, we formulate the meta-learning process as a compositional image editing task that learns to imitate a certain visual effect and derive the corresponding internal representation. Such a generative process can help instantiate the underlying figure-ground notion and enables the system to accomplish the intended image segmentation. Whereas existing generative methods are mostly tailored to image synthesis or style transfer, our approach offers a flexible learning mechanism to model a general concept of figure-ground segmentation from unorganized images that have no explicit pixel-level annotations. We validate our approach via extensive experiments on six datasets to demonstrate that the proposed model can be end-to-end trained without ground-truth pixel labeling yet outperforms the existing methods of unsupervised segmentation tasks.
It is well believed that video captioning is a fundamental but challenging task in both computer vision and artificial intelligence fields. The prevalent approach is to map an input video to a variable-length output sentence in a sequence to sequence manner via Recurrent Neural Network (RNN). Nevertheless, the training of RNN still suffers to some degree from vanishing/exploding gradient problem, making the optimization difficult. Moreover, the inherently recurrent dependency in RNN prevents parallelization within a sequence during training and therefore limits the computations. In this paper, we present a novel design — Temporal Deformable Convolutional Encoder-Decoder Networks (dubbed as TDConvED) that fully employ convolutions in both encoder and decoder networks for video captioning. Technically, we exploit convolutional block structures that compute intermediate states of a fixed number of inputs and stack several blocks to capture long-term relationships. The structure in encoder is further equipped with temporal deformable convolution to enable free-form deformation of temporal sampling. Our model also capitalizes on temporal attention mechanism for sentence generation. Extensive experiments are conducted on both MSVD and MSR-VTT video captioning datasets, and superior results are reported when comparing to conventional RNN-based encoder-decoder techniques. More remarkably, TDConvED increases CIDEr-D performance from 58.8% to 67.2% on MSVD.
In this paper, we consider the task of natural language video localization (NLVL): given an untrimmed video and a natural language description, the goal is to localize a segment in the video which semantically corresponds to the given natural language description. We propose a localizing network (LNet), working in an end-to-end fashion, to tackle the NLVL task. We first match the natural sentence and video sequence by cross-gated attended recurrent networks to exploit their fine-grained interactions and generate a sentence-aware video representation. A self interactor is proposed to perform crossframe matching, which dynamically encodes and aggregates the matching evidences. Finally, a boundary model is proposed to locate the positions of video segments corresponding to the natural sentence description by predicting the starting and ending points of the segment. Extensive experiments conducted on the public TACoS and DiDeMo datasets demonstrate that our proposed model performs effectively and efficiently against the state-of-the-art approaches.
Quantization has been widely adopted for large-scale multimedia retrieval due to its effectiveness of coding highdimensional data. Deep quantization models have been demonstrated to achieve the state-of-the-art retrieval accuracy. However, training the deep models given a large-scale database is highly time-consuming as a large amount of parameters are involved. Existing deep quantization methods often sample only a subset from the database for training, which may end up with unsatisfactory retrieval performance as a large portion of label information is discarded. To alleviate this problem, we propose a novel model called Similarity Preserving Deep Asymmetric Quantization (SPDAQ) which can directly learn the compact binary codes and quantization codebooks for all the items in the database efficiently. To do that, SPDAQ makes use of an image subset as well as the label information of all the database items so the image subset items and the database items are mapped to two different but correlated distributions, where the label similarity can be well preserved. An efficient optimization algorithm is proposed for the learning. Extensive experiments conducted on four widely-used benchmark datasets demonstrate the superiority of our proposed SPDAQ model.
Sequence-to-sequence models incorporated with attention mechanism have shown promising improvements on video captioning. While there is rich information both inside and between frames, spatial attention is rarely explored and motion information is usually handled by 3D-CNNs as just another modality for fusion. On the other hand, researches about human perception suggest that apparent motion can attract attention. Motivated by this, we aim to learn spatial attention on video frames under the guidance of motion information for caption generation. We present a novel video captioning framework by utilizing Motion Guided Spatial Attention (MGSA). The proposed MGSA exploits the motion between video frames by learning spatial attention from stacked optical flow images with a custom CNN. To further relate the spatial attention maps of video frames, we designed a Gated Attention Recurrent Unit (GARU) to adaptively incorporate previous attention maps. The whole framework can be trained in an end-to-end manner. We evaluate our approach on two benchmark datasets, MSVD and MSR-VTT. The experiments show that our designed model can generate better video representation and state of the art results are obtained under popular evaluation metrics such as BLEU@4, CIDEr, and METEOR.
This paper presents an efficient algorithm to tackle temporal localization of activities in videos via sentence queries. The task differs from traditional action localization in three aspects: (1) Activities are combinations of various kinds of actions and may span a long period of time. (2) Sentence queries are not limited to a predefined list of classes. (3) The videos usually contain multiple different activity instances. Traditional proposal-based approaches for action localization that only consider the class-agnostic “actionness” of video snippets are insufficient to tackle this task. We propose a novel Semantic Activity Proposal (SAP) which integrates the semantic information of sentence queries into the proposal generation process to get discriminative activity proposals. Visual and semantic information are jointly utilized for proposal ranking and refinement. We evaluate our algorithm on the TACoS dataset and the Charades-STA dataset. Experimental results show that our algorithm outperforms existing methods on both datasets, and at the same time reduces the number of proposals by a factor of at least 10.
Bilingual lexicon induction, translating words from the source language to the target language, is a long-standing natural language processing task. Recent endeavors prove that it is promising to employ images as pivot to learn the lexicon induction without reliance on parallel corpora. However, these vision-based approaches simply associate words with entire images, which are constrained to translate concrete words and require object-centered images. We humans can understand words better when they are within a sentence with context. Therefore, in this paper, we propose to utilize images and their associated captions to address the limitations of previous approaches. We propose a multi-lingual caption model trained with different mono-lingual multimodal data to map words in different languages into joint spaces. Two types of word representation are induced from the multi-lingual caption model: linguistic features and localized visual features. The linguistic feature is learned from the sentence contexts with visual semantic constraints, which is beneficial to learn translation for words that are less visual-relevant. The localized visual feature is attended to the region in the image that correlates to the word, so that it alleviates the image restriction for salient visual representation. The two types of features are complementary for word translation. Experimental results on multiple language pairs demonstrate the effectiveness of our proposed method, which substantially outperforms previous vision-based approaches without using any parallel sentences or supervision of seed word pairs.
Person re-identification (re-ID) solves the task of matching images across cameras and is among the research topics in vision community. Since query images in real-world scenarios might suffer from resolution loss, how to solve the resolution mismatch problem during person re-ID becomes a practical problem. Instead of applying separate image super-resolution models, we propose a novel network architecture of Resolution Adaptation and re-Identification Network (RAIN) to solve cross-resolution person re-ID. Advancing the strategy of adversarial learning, we aim at extracting resolution-invariant representations for re-ID, while the proposed model is learned in an end-to-end training fashion. Our experiments confirm that the use of our model can recognize low-resolution query images, even if the resolution is not seen during training. Moreover, the extension of our model for semi-supervised re-ID further confirms the scalability of our proposed method for real-world scenarios and applications.
In real-world applications, commercial off-the-shelf systems are utilized for performing automated facial analysis including face recognition, emotion recognition, and attribute prediction. However, a majority of these commercial systems act as black boxes due to the inaccessibility of the model parameters which makes it challenging to fine-tune the models for specific applications. Stimulated by the advances in adversarial perturbations, this research proposes the concept of Data Fine-tuning to improve the classification accuracy of a given model without changing the parameters of the model. This is accomplished by modeling it as data (image) perturbation problem. A small amount of “noise” is added to the input with the objective of minimizing the classification loss without affecting the (visual) appearance. Experiments performed on three publicly available datasets LFW, CelebA, and MUCT, demonstrate the effectiveness of the proposed concept.
High performance face detection remains a very challenging problem, especially when there exists many tiny faces. This paper presents a novel single-shot face detector, named Selective Refinement Network (SRN), which introduces novel twostep classification and regression operations selectively into an anchor-based face detector to reduce false positives and improve location accuracy simultaneously. In particular, the SRN consists of two modules: the Selective Two-step Classification (STC) module and the Selective Two-step Regression (STR) module. The STC aims to filter out most simple negative anchors from low level detection layers to reduce the search space for the subsequent classifier, while the STR is designed to coarsely adjust the locations and sizes of anchors from high level detection layers to provide better initialization for the subsequent regressor. Moreover, we design a Receptive Field Enhancement (RFE) block to provide more diverse receptive field, which helps to better capture faces in some extreme poses. As a consequence, the proposed SRN detector achieves state-of-the-art performance on all the widely used face detection benchmarks, including AFW, PASCAL face, FDDB, and WIDER FACE datasets. Codes will be released to facilitate further studies on the face detection problem.
Heterogeneous Face Recognition (HFR) is a challenging task due to large modality discrepancy as well as insufficient training images in certain modalities. In this paper, we propose a new two-branch network architecture, termed as Residual Compensation Networks (RCN), to learn separated features for different modalities in HFR. The RCN incorporates a residual compensation (RC) module and a modality discrepancy loss (MD loss) into traditional convolutional neural networks. The RC module reduces modal discrepancy by adding compensation to one of the modalities so that its representation can be close to the other modality. The MD loss alleviates modal discrepancy by minimizing the cosine distance between different modalities. In addition, we explore different architectures and positions for the RC module, and evaluate different transfer learning strategies for HFR. Extensive experiments on IIIT-D Viewed Sketch, Forensic Sketch, CASIA NIR-VIS 2.0 and CUHK NIR-VIS show that our RCN outperforms other state-of-the-art methods significantly.
Deep learning based methods have achieved remarkable progress in action recognition. Existing works mainly focus on designing novel deep architectures to achieve video representations learning for action recognition. Most methods treat sampled frames equally and average all the frame-level predictions at the testing stage. However, within a video, discriminative actions may occur sparsely in a few frames and most other frames are irrelevant to the ground truth and may even lead to a wrong prediction. As a result, we think that the strategy of selecting relevant frames would be a further important key to enhance the existing deep learning based action recognition. In this paper, we propose an attentionaware sampling method for action recognition, which aims to discard the irrelevant and misleading frames and preserve the most discriminative frames. We formulate the process of mining key frames from videos as a Markov decision process and train the attention agent through deep reinforcement learning without extra labels. The agent takes features and predictions from the baseline model as input and generates importance scores for all frames. Moreover, our approach is extensible, which can be applied to different existing deep learning based action recognition models. We achieve very competitive action recognition performance on two widely used action recognition datasets.
In the monochrome-color dual-lens system, the gray image captured by the monochrome camera has better quality than the color image from the color camera, but does not have color information. To get high-quality color images, it is desired to colorize the gray image with the color image as reference. Related works usually use hand-crafted methods to search for the best-matching pixel in the reference image for each pixel in the input gray image, and copy the color of the best-matching pixel as the result. We propose a novel deep convolution network to solve the colorization problem in an end-to-end way. Based on our observation that, for each pixel in the input image, there usually exist multiple pixels in the reference image that have the correct colors, our method performs weighted average of colors of the candidate pixels in the reference image to utilize more candidate pixels with correct colors. The weight values between pixels in the input image and the reference image are obtained by learning a weight volume using deep feature representations, where an attention operation is proposed to focus on more useful candidate pixels and a 3-D regulation is performed to learn with context information. In addition, to correct wrongly colorized pixels in occlusion regions, we propose a color residue joint learning module to correct the colorization result with the input gray image as guidance. We evaluate our method on the Scene Flow, Cityscapes, Middlebury, and Sintel datasets. Experimental results show that our method largely outperforms the state-of-the-art methods.
Predicting future frames in videos has become a promising direction of research for both computer vision and robot learning communities. The core of this problem involves moving object capture and future motion prediction. While object capture specifies which objects are moving in videos, motion prediction describes their future dynamics. Motivated by this analysis, we propose a Cubic Long Short-Term Memory (CubicLSTM) unit for video prediction. CubicLSTM consists of three branches, i.e., a spatial branch for capturing moving objects, a temporal branch for processing motions, and an output branch for combining the first two branches to generate predicted frames. Stacking multiple CubicLSTM units along the spatial branch and output branch, and then evolving along the temporal branch can form a cubic recurrent neural network (CubicRNN). Experiment shows that CubicRNN produces more accurate video predictions than prior methods on both synthetic and real-world datasets.
Automatically generating natural language description for video is an extremely complicated and challenging task. To tackle the obstacles of traditional LSTM-based model for video captioning, we propose a novel architecture to generate the optimal descriptions for videos, which focuses on constructing a new network structure that can generate sentences superior to the basic model with LSTM, and establishing special attention mechanisms that can provide more useful visual information for caption generation. This scheme discards the traditional LSTM, and exploits the fully convolutional network with coarse-to-fine and inherited attention designed according to the characteristics of fully convolutional structure. Our model cannot only outperform the basic LSTM-based model, but also achieve the comparable performance with those of state-of-the-art methods