Designing accurate and efficient ConvNets for mobile devices is challenging because the design space is combinatorially large. Due to this, previous neural architecture search (NAS) methods are ...computationally expensive. ConvNet architecture optimality depends on factors such as input resolution and target devices. However, existing approaches are too resource demanding for case-by-case redesigns. Also, previous work focuses primarily on reducing FLOPs, but FLOP count does not always reflect actual latency. To address these, we propose a differentiable neural architecture search (DNAS) framework that uses gradient-based methods to optimize ConvNet architectures, avoiding enumerating and training individual architectures separately as in previous methods. FBNets (Facebook-Berkeley-Nets), a family of models discovered by DNAS surpass state-of-the-art models both designed manually and generated automatically. FBNet-B achieves 74.1% top-1 accuracy on ImageNet with 295M FLOPs and 23.1 ms latency on a Samsung S8 phone, 2.4x smaller and 1.5x faster than MobileNetV2-1.3 with similar accuracy. Despite higher accuracy and lower latency than MnasNet, we estimate FBNet-B's search cost is 420x smaller than MnasNet's, at only 216 GPU-hours. Searched for different resolutions and channel sizes, FBNets achieve 1.5% to 6.4% higher accuracy than MobileNetV2. The smallest FBNet achieves 50.2% accuracy and 2.9 ms latency (345 frames per second) on a Samsung S8. Over a Samsung-optimized FBNet, the iPhone-X-optimized model achieves a 1.4x speedup on an iPhone X. FBNet models are open-sourced at https://github. com/facebookresearch/mobile-vision.
Semantic Amodal Segmentation Yan Zhu; Yuandong Tian; Metaxas, Dimitris ...
2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR),
2017-July
Conference Proceeding
Common visual recognition tasks such as classification, object detection, and semantic segmentation are rapidly reaching maturity, and given the recent rate of progress, it is not unreasonable to ...conjecture that techniques for many of these problems will approach human levels of performance in the next few years. In this paper we look to the future: what is the next frontier in visual recognition? We offer one possible answer to this question. We propose a detailed image annotation that captures information beyond the visible pixels and requires complex reasoning about full scene structure. Specifically, we create an amodal segmentation of each image: the full extent of each region is marked, not just the visible pixels. Annotators outline and name all salient regions in the image and specify a partial depth order. The result is a rich scene structure, including visible and occluded portions of each region, figure-ground edge information, semantic labels, and object overlap. We create two datasets for semantic amodal segmentation. First, we label 500 images in the BSDS dataset with multiple annotators per image, allowing us to study the statistics of human annotations. We show that the proposed full scene annotation is surprisingly consistent between annotators, including for regions and edges. Second, we annotate 5000 images from COCO. This larger dataset allows us to explore a number of algorithmic ideas for amodal segmentation and depth ordering. We introduce novel metrics for these tasks, and along with our strong baselines, define concrete new challenges for the community.
Image alignment in the presence of non-rigid distortions is a challenging task. Typically, this involves estimating the parameters of a dense deformation field that warps a distorted image back to ...its undistorted template. Generative approaches based on parameter optimization such as Lucas-Kanade can get trapped within local minima. On the other hand, discriminative approaches like nearest-neighbor require a large number of training samples that grows exponentially with respect to the dimension of the parameter space, and polynomially with the desired accuracy 1/
ϵ
. In this work, we develop a novel data-driven iterative algorithm that combines the best of both generative and discriminative approaches. For this, we introduce the notion of a “pull-back” operation that enables us to predict the parameters of the test image using training samples that are not in its neighborhood (not
ϵ
-close) in the parameter space. We prove that our algorithm converges to the global optimum using a significantly lower number of training samples that grows only
logarithmically
with the desired accuracy. We analyze the behavior of our algorithm extensively using synthetic data and demonstrate successful results on experiments with complex deformations due to water and clothing.
Full text
Available for:
CEKLJ, DOBA, EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, IZUM, KILJ, KISLJ, MFDPS, NLZOH, NUK, OILJ, PILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, SIK, UILJ, UKNU, UL, UM, UPUK, VKSCE, ZAGLJ
We introduce a new memory architecture, Bayesian Relational Memory (BRM), to improve the generalization ability for semantic visual navigation agents in unseen environments, where an agent is given a ...semantic target to navigate towards. BRM takes the form of a probabilistic relation graph over semantic entities (e.g., room types), which allows (1) capturing the layout prior from training environments, i.e., prior knowledge, (2) estimating posterior layout at test time, i.e., memory update, and (3) efficient planning for navigation, altogether. We develop a BRM agent consisting of a BRM module for producing sub-goals and a goal-conditioned locomotion module for control. When testing in unseen environments, the BRM agent outperforms baselines that do not explicitly utilize the probabilistic relational memory structure.
Human pose estimation requires a versatile yet well-constrained spatial model for grouping locally ambiguous parts together to produce a globally consistent hypothesis. Previous works either use ...local deformable models deviating from a certain template, or use a global mixture representation in the pose space. In this paper, we propose a new hierarchical spatial model that can capture an exponential number of poses with a compact mixture representation on each part. Using latent nodes, it can represent high-order spatial relationship among parts with exact inference. Different from recent hierarchical models that associate each latent node to a mixture of appearance templates (like HoG), we use the hierarchical structure as a pure spatial prior avoiding the large and often confounding appearance space. We verify the effectiveness of this model in three ways. First, samples representing human-like poses can be drawn from our model, showing its ability to capture high-order dependencies of parts. Second, our model achieves accurate reconstruction of unseen poses compared to a nearest neighbor pose representation. Finally, our model achieves state-of-art performance on three challenging datasets, and substantially outperforms recent hierarchical models.
Hierarchical learning has been successful at learning generalizable locomotion skills on walking robots in a sample-efficient manner. However, the low-dimensional "latent" action used to communicate ...between two layers of the hierarchy is typically user-designed. In this letter, we present a fully-learned hierarchical framework, that is capable of jointly learning the low-level controller and the high-level latent action space. Once this latent space is learned, we plan over continuous latent actions in a model-predictive control fashion, using a learned high-level dynamics model. This framework generalizes to multiple robots, and we present results on a Daisy hexapod simulation, A1 quadruped simulation, and Daisy robot hardware. We compare a range of learned hierarchical approaches from literature, and show that our framework outperforms baselines on multiple tasks and two simulations. In addition to learning approaches, we also compare to inverse-kinematics (IK) acting on desired robot motion, and show that our fully-learned framework outperforms IK in adverse settings on both A1 and Daisy simulations. On hardware, we show the Daisy hexapod achieve multiple locomotion tasks, in an unstructured outdoor setting, with only 2000 hardware samples, reinforcing the robustness and sample-efficiency of our approach.
Real-world surfaces such as clothing, water and human body deform in complex ways. The image distortions observed are high-dimensional and non-linear, making it hard to estimate these deformations ...accurately. The recent data-driven descent approach applies Nearest Neighbor estimators iteratively on a particular distribution of training samples to obtain a globally optimal and dense deformation field between a template and a distorted image. In this work, we develop a hierarchical structure for the Nearest Neighbor estimators, each of which can have only a local image support. We demonstrate in both theory and practice that this algorithm has several advantages over the non-hierarchical version: it guarantees global optimality with significantly fewer training samples, is several orders faster, provides a metric to decide whether a given image is ``hard'' (or ``easy'') requiring more (or less) samples, and can handle more complex scenes that include both global motion and local deformation. The proposed algorithm successfully tracks a broad range of non-rigid scenes including water, clothing, and medical images, and compares favorably against several other deformation estimation and tracking approaches that do not provide optimality guarantees.
Projectors are increasingly being used as light-sources in computer vision applications. In several applications, they are modeled as point light sources, thus ignoring the effects of illumination ...defocus. In addition, most active vision techniques assume that a scene point is illuminated only directly by the light source, thus ignoring global light transport effects. Since both defocus and global illumination co-occur in virtually all scenes illuminated by projectors, ignoring them can result in strong, systematic biases in the recovered scene properties. To make computer vision techniques work for general real world scenes, it is thus important to account for both these effects.
In this paper, we study the interplay between defocused illumination and global light transport. We show that both these seemingly disparate effects can be expressed as low pass filters on the incident illumination. Using this observation, we derive an invariant between the two effects, which can be used to separate the two. This is directly useful in scenarios where limited depth-of-field devices (such as projectors) are used to illuminate scenes with global light transport and significant depth variations. We show applications in two scenarios: (a) accurate depth recovery in the presence of global light transport, and (b) factoring out the effects of illumination defocus for correct direct-global component separation. We demonstrate our approach using scenes with complex shapes, reflectance properties, textures and translucencies.
Full text
Available for:
CEKLJ, DOBA, EMUNI, FIS, FZAB, GEOZS, GIS, IJS, IMTLJ, IZUM, KILJ, KISLJ, MFDPS, NLZOH, NUK, OILJ, PILJ, PNG, SAZU, SBCE, SBJE, SBMB, SBNM, SIK, UILJ, UKNU, UL, UM, UPUK, VKSCE, ZAGLJ
Distortions in images of documents, such as the pages of books, adversely affect the performance of optical character recognition (OCR) systems. Removing such distortions requires the 3D deformation ...of the document that is often measured using special and precisely calibrated hardware (stereo, laser range scanning or structured light). In this paper, we introduce a new approach that automatically reconstructs the 3D shape and rectifies a deformed text document from a single image. We first estimate the 2D distortion grid in an image by exploiting the line structure and stroke statistics in text documents. This approach does not rely on more noise-sensitive operations such as image binarization and character segmentation. The regularity in the text pattern is used to constrain the 2D distortion grid to be a perspective projection of a 3D parallelogram mesh. Based on this constraint, we present a new shape-from-texture method that computes the 3D deformation up to a scale factor using SVD. Unlike previous work, this formulation imposes no restrictions on the shape (e.g., a developable surface). The estimated shape is then used to remove both geometric distortions and photometric (shading) effects in the image. We demonstrate our techniques on documents containing a variety of languages, fonts and sizes.
Neural Architecture Search (NAS) has emerged as a promising technique for automatic neural network design. However, existing MCTS based NAS approaches often utilize manually designed action space, ...which is not directly related to the performance metric to be optimized (e.g., accuracy), leading to sample-inefficient explorations of architectures. To improve the sample efficiency, this paper proposes Latent Action Neural Architecture Search (LaNAS), which learns actions to recursively partition the search space into good or bad regions that contain networks with similar performance metrics. During the search phase, as different action sequences lead to regions with different performance, the search efficiency can be significantly improved by biasing towards the good regions. On three NAS tasks, empirical results demonstrate that LaNAS is at least an order more sample efficient than baseline methods including evolutionary algorithms, Bayesian optimizations, and random search. When applied in practice, both one-shot and regular LaNAS consistently outperform existing results. Particularly, LaNAS achieves 99.0 percent accuracy on CIFAR-10 and 80.8 percent top1 accuracy at 600 MFLOPS on ImageNet in only 800 samples, significantly outperforming AmoebaNet with <inline-formula><tex-math notation="LaTeX">33\times</tex-math> <mml:math><mml:mrow><mml:mn>33</mml:mn><mml:mo>×</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="wang-ieq1-3071343.gif"/> </inline-formula> fewer samples. Our code is publicly available at https://github.com/facebookresearch/LaMCTS .