Subspace clustering methods have been widely studied recently. When the inputs are 2-dimensional (2D) data, existing subspace clustering methods usually convert them into vectors, which severely damages inherent structures and relationships from original data. In this paper, we propose a novel subspace clustering method for 2D data. It directly uses 2D data as inputs such that the learning of representations benefits from inherent structures and relationships of the data. It simultaneously seeks image projection and representation coefficients such that they mutually enhance each other and lead to powerful data representations. An efficient algorithm is developed to solve the proposed objective function with provable decreasing and convergence property. Extensive experimental results verify the effectiveness of the new method.
In this paper we present a Transformer-Transducer model architecture and a training technique to unify streaming and non-streaming speech recognition models into one model. The model is composed of a stack of transformer layers for audio encoding with no lookahead or right context and an additional stack of transformer layers on top trained with variable right context. In inference time, the context length for the variable context layers can be changed to trade off the latency and the accuracy of the model. We also show that we can run this model in a Y-model architecture with the top layers running in parallel in low latency and high latency modes. This allows us to have streaming speech recognition results with limited latency and delayed speech recognition results with large improvements in accuracy (20% relative improvement for voice-search task). We show that with limited right context (1-2 seconds of audio) and small additional latency (50-100 milliseconds) at the end of decoding, we can achieve similar accuracy with models using unlimited audio right context. We also present optimizations for audio and label encoders to speed up the inference in streaming and non-streaming speech decoding.
Discrete optimization is one of the most intractable problems in deep hashing. Previous methods usually mitigate this problem by binary approximation, substituting binary codes for real-values via activation functions or regularizations. However, such approximation leads to uncertainty between real-values and binary ones, degrading retrieval performance. In this paper, we propose a novel Deep Momentum Uncertainty Hashing (DMUH). It explicitly estimates the uncertainty during training and leverages the uncertainty information to guide the approximation process. Specifically, we model \emph{bit-level uncertainty} via measuring the discrepancy between the output of a hashing network and that of a momentum-updated network. The discrepancy of each bit indicates the uncertainty of the hashing network to the approximate output of that bit. Meanwhile, the mean discrepancy of all bits in a hashing code can be regarded as \emph{image-level uncertainty}. It embodies the uncertainty of the hashing network to the corresponding input image. The hashing bit and the image with higher uncertainty are paid more attention during optimization. To the best of our knowledge, this is the first work to study the uncertainty in hashing bits. Extensive experiments are conducted on four datasets to verify the superiority of our method, including CIFAR-10, NUS-WIDE, MS-COCO, and a million-scale dataset Clothing1M. Our method achieves best performance on all datasets and surpasses existing state-of-the-arts by a large margin, especially on Clothing1M.
We present AutoPose, a novel neural architecture search(NAS) framework that is capable of automatically discovering multiple parallel branches of cross-scale connections towards accurate and high-resolution 2D human pose estimation. Recently, high-performance hand-crafted convolutional networks for pose estimation show growing demands on multi-scale fusion and high-resolution representations. However, current NAS works exhibit limited flexibility on scale searching, they dominantly adopt simplified search spaces of single-branch architectures. Such simplification limits the fusion of information at different scales and fails to maintain high-resolution representations. The presentedAutoPose framework is able to search for multi-branch scales and network depth, in addition to the cell-level microstructure. Motivated by the search space, a novel bi-level optimization method is presented, where the network-level architecture is searched via reinforcement learning, and the cell-level search is conducted by the gradient-based method. Within 2.5 GPU days, AutoPose is able to find very competitive architectures on the MS COCO dataset, that are also transferable to the MPII dataset. Our code is available at https://github.com/VITA-Group/AutoPose.
Despite the previous success of object analysis, detecting and segmenting a large number of object categories with a long-tailed data distribution remains a challenging problem and is less investigated. For a large-vocabulary classifier, the chance of obtaining noisy logits is much higher, which can easily lead to a wrong recognition. In this paper, we exploit prior knowledge of the relations among object categories to cluster fine-grained classes into coarser parent classes, and construct a classification tree that is responsible for parsing an object instance into a fine-grained category via its parent class. In the classification tree, as the number of parent class nodes are significantly less, their logits are less noisy and can be utilized to suppress the wrong/noisy logits existed in the fine-grained class nodes. As the way to construct the parent class is not unique, we further build multiple trees to form a classification forest where each tree contributes its vote to the fine-grained classification. To alleviate the imbalanced learning caused by the long-tail phenomena, we propose a simple yet effective resampling method, NMS Resampling, to re-balance the data distribution. Our method, termed as Forest R-CNN, can serve as a plug-and-play module being applied to most object recognition models for recognizing more than 1000 categories. Extensive experiments are performed on the large vocabulary dataset LVIS. Compared with the Mask R-CNN baseline, the Forest R-CNN significantly boosts the performance with 11.5% and 3.9% AP improvements on the rare categories and overall categories, respectively. Moreover, we achieve state-of-the-art results on the LVIS dataset. Code is available at https://github.com/JialianW/Forest_RCNN.
General movement assessment (GMA) of infant movement videos (IMVs) is an effective method for the early detection of cerebral palsy (CP) in infants. Automated body parsing is a crucial step towards computer-aided GMA, in which infant body parts are segmented and tracked over time for movement analysis. However, acquiring fully annotated data for video-based body parsing is particularly expensive due to the large number of frames in IMVs. In this paper, we propose a semi-supervised body parsing model, termed SiamParseNet (SPN), to jointly learn single frame body parsing and label propagation between frames in a semi-supervised fashion. The Siamese-structured SPN consists of a shared feature encoder, followed by two separate branches: one for intra-frame body parts segmentation, and one for inter-frame label propagation. The two branches are trained jointly, taking pairs of frames from the same videos as their input. An adaptive training process is proposed that alternates training modes between using input pairs of only labeled frames and using inputs of both labeled and unlabeled frames. During testing, we employ a multi-source inference mechanism, where the final result for a test frame is either obtained via the segmentation branch or via propagation from a nearby key frame. We conduct extensive experiments on a partially-labeled IMV dataset where SPN outperforms all prior arts, demonstrating the effectiveness of our proposed method.
In this paper we study Meta learning of Gaussian graphical models. In our setup, each task has a different true precision matrix, each with a possibly different support (i.e., set of edges in the graph). We assume that the union of the supports of all the true precision matrices (i.e., the true support union) is small in size, which relates to sparse graphs. We propose to pool all the samples from different tasks, and estimate a single precision matrix by $\ell_1$-regularized maximum likelihood estimation. We show that with high probability, the support of the estimated single precision matrix is equal to the true support union, provided a sufficient number of samples per task $n \in O((\log N)/K)$, for $N$ nodes and $K$ tasks. That is, one requires less samples per task when more tasks are available. We prove a matching information-theoretic lower bound for the necessary number of samples, which is $n \in \Omega((\log N)/K)$, and thus, our algorithm is minimax optimal. Synthetic experiments validate our theory.
Deep neural networks achieve remarkable performance in many computer vision tasks. Most state-of-the-art (SOTA) semantic segmentation and object detection approaches reuse neural network architectures designed for image classification as the backbone, commonly pre-trained on ImageNet. However, performance gains can be achieved by designing network architectures specifically for detection and segmentation, as shown by recent neural architecture search (NAS) research for detection and segmentation. One major challenge though is that ImageNet pre-training of the search space representation (a.k.a. super network) or the searched networks incurs huge computational cost. In this paper, we propose a Fast Network Adaptation (FNA++) method, which can adapt both the architecture and parameters of a seed network (e.g. an ImageNet pre-trained network) to become a network with different depths, widths, or kernel sizes via a parameter remapping technique, making it possible to use NAS for segmentation/detection tasks a lot more efficiently. In our experiments, we conduct FNA++ on MobileNetV2 to obtain new networks for semantic segmentation, object detection, and human pose estimation that clearly outperform existing networks designed both manually and by NAS. We also implement FNA++ on ResNets and NAS networks, which demonstrates a great generalization ability. The total computation cost of FNA++ is significantly less than SOTA segmentation/detection NAS approaches: 1737x less than DPC, 6.8x less than Auto-DeepLab, and 8.0x less than DetNAS. The code will be released at https://github.com/JaminFong/FNA.
Automated tongue image segmentation in tongue images is a challenging task for two reasons: 1) there are many pathological details on the tongue surface, which affect the extraction of the boundary; 2) the shapes of the tongues captured from various persons (with different diseases) are quite different. To deal with the challenge, a novel end-to-end Boundary Guidance Hierarchical Network (BGHNet) with a new hybrid loss is proposed in this paper. In the new approach, firstly Context Feature Encoder Module (CFEM) is built upon the bottomup pathway to confront with the shrinkage of the receptive field. Secondly, a novel hierarchical recurrent feature fusion module (HRFFM) is adopt to progressively and hierarchically refine object maps to recover image details by integrating local context information. Finally, the proposed hybrid loss in a four hierarchy-pixel, patch, map and boundary guides the network to effectively segment the tongue regions and accurate tongue boundaries. BGHNet is applied to a set of tongue images. The experimental results suggest that the proposed approach can achieve the latest tongue segmentation performance. And in the meantime, the lightweight network contains only 15.45M parameters and performs only 11.22GFLOPS.
This paper studies a new problem, namely active lighting recurrence (ALR) that physically relocalizes a light source to reproduce the lighting condition from single reference image for a same scene, which may suffer from fine-grained changes during twice observations. ALR is of great importance for fine-grained visual inspection and change detection, because some phenomena or minute changes can only be clearly observed under particular lighting conditions. Therefore, effective ALR should be able to online navigate a light source toward the target pose, which is challenging due to the complexity and diversity of real-world lighting and imaging processes. To this end, we propose to use the simple parallel lighting as an analogy model and based on Lambertian law to compose an instant navigation ball for this purpose. We theoretically prove the feasibility, i.e., equivalence and convergence, of this ALR approach for realistic near point light source and small near surface light source. Besides, we also theoretically prove the invariance of our ALR approach to the ambiguity of normal and lighting decomposition. The effectiveness and superiority of the proposed approach have been verified by both extensive quantitative experiments and challenging real-world tasks on fine-grained change detection of cultural heritages. We also validate the generality of our approach to non-Lambertian scenes.