AI has seen remarkable progress in recent years, due to a switch from hand-designed shallow representations, to learned deep representations. While these methods excel with plentiful training data, they are still far from the human ability to learn concepts from just a few examples by reusing previously learned conceptual knowledge in new contexts. We argue that this gap might come from a fundamental misalignment between human and typical AI representations: while the former are grounded in rich sensorimotor experience, the latter are typically passive and limited to a few modalities such as vision and text. We take a step towards closing this gap by proposing an interactive, behavior-based model that represents concepts using sensorimotor contingencies grounded in an agent's experience. On a novel conceptual learning and benchmark suite, we demonstrate that conceptually meaningful behaviors can be learned, given supervision via training curricula.

Convolutional neural networks (CNNs) have shown great success in computer vision, approaching human-level performance when trained for specific tasks via application-specific loss functions. In this paper, we propose a method for augmenting and training CNNs so that their learned features are compositional. It encourages networks to form representations that disentangle objects from their surroundings and from each other, thereby promoting better generalization. Our method is agnostic to the specific details of the underlying CNN to which it is applied and can in principle be used with any CNN. As we show in our experiments, the learned representations lead to feature activations that are more localized and improve performance over non-compositional baselines in object recognition tasks.

Fine-grained recognition concerns categorization at sub-ordinate levels, where the distinction between object classes is highly local. Compared to basic level recognition, fine-grained categorization can be more challenging as there are in general less data and fewer discriminative features. This necessitates the use of a stronger prior for feature selection. In this work, we include humans in the loop to help computers select discriminative features. We introduce a novel online game called “Bubbles” that reveals discriminative features humans use. The player's goal is to identify the category of a heavily blurred image. During the game, the player can choose to reveal full details of circular regions (“bubbles”), with a certain penalty. With proper setup the game generates discriminative bubbles with assured quality. We next propose the “BubbleBank” representation that uses the human selected bubbles to improve machine recognition performance. Finally, we demonstrate how to extend BubbleBank to a view-invariant 3D representation. Experiments demonstrate that our approach yields large improvements over the previous state of the art on challenging benchmarks.

In this paper, a method for the visualization of the population of an evolutionary multi-objective optimization (EMO) algorithm is presented. The main characteristic of this approach is the preservation of Paretodominance relations among the individuals as good as possible. It will be shown that in general, a Paretodominance preserving mapping from higher- to lowerdimensional spaces does not exist. Thus, the demand is to find a mapping with as few wrongly indicated dominance relations as possible, which gives one more objective in addition to other mapping objectives like preserving nearest neighbor relations. Therefore, such a mapping poses a multi-objective optimization problem by itself, which is also handled by an EMO algorithm (NSGA-II in this case). The resulting mappings are shown for the run of a NSGA-II version on the 15 objective DTLZ2 problem as an example. From such plots, some insights into evolutionary dynamics can be obtained.

It this section, we provide qualitative examples and plots for the experiment “2D-3D Matching as an Object Detector” (Sect. 6.2 in the main paper). To recapitulate, we run our ensemble of NZ-WHO templates on the 3D Object Classes dataset [2], without the finetuning stage. Fig. 1 gives the corresponding detection average precision, average viewpoint precision, viewpoint confusion matrix and mean precision in pose estimation results. Specifically, we followed the detection and viewpoint estimation criteria of [4] where a detection is correct iff intersection over union is at least 0.5 and viewpoint estimation is correct iff detection is correct and azimuth of the viewpoint prediction falls into the correct viewpoint bin. Fig. 2 shows successful detection and viewpoint estimation results for car, Fig. 3 for bicycle. Fig. 4 and Fig. 5 show failure cases, which are mostly due to confused front and back views for cars, and slanted bicycle poses.

A large body of recent work on object detection has focused on exploiting 3D CAD model databases to improve detection performance. Many of these approaches work by aligning exact 3D models to images using templates generated from renderings of the 3D models at a set of discrete viewpoints. However, the training procedures for these approaches are computationally expensive and require gigabytes of memory and storage, while the viewpoint discretization hampers pose estimation performance.We propose an efficient method for synthesizing templates from 3D models that runs on the fly - that is, it quickly produces detectors for an arbitrary viewpoint of a 3D model without expensive dataset-dependent training or template storage. Given a 3D model and an arbitrary continuous detection viewpoint, our method synthesizes a discriminative template by extracting features from a rendered view of the object and decorrelating spatial dependences among the features. Our decorrelation procedure relies on a gradient-based algorithm that is more numerically stable than standard decomposition-based procedures, and we efficiently search for candidate detections by computing FFT-based template convolutions. Due to the speed of our template synthesis procedure, we are able to perform joint optimization of scale, translation, continuous rotation, and focal length using Metropolis-Hastings algorithm. We provide an efficient GPU implementation of our algorithm, and we validate its performance on 3D Object Classes and PASCAL3D+ datasets.

Object class detection has been a synonym for 2D bounding box localization for the longest time, fueled by the success of powerful statistical learning techniques, combined with robust image representations. Only recently, there has been a growing interest in revisiting the promise of computer vision from the early days: to precisely delineate the contents of a visual scene, object by object, in 3D. In this paper, we draw from recent advances in object detection and 2D-3D object lifting in order to design an object class detector that is particularly tailored towards 3D object class detection. Our 3D object class detection method consists of several stages gradually enriching the object detection output with object viewpoint, keypoints and 3D shape estimates. Following careful design, in each stage it constantly improves the performance and achieves state-ofthe-art performance in simultaneous 2D bounding box and viewpoint estimation on the challenging Pascal3D+ dataset.

As objects are inherently 3D, they have been modeled in 3D in the early days of computer vision. Due to the ambiguities arising from mapping 2D features to 3D models, 3D object representations have been neglected and 2D feature-based models are the predominant paradigm in object detection nowadays. While such models have achieved outstanding bounding box detection performance, they come with limited expressiveness, as they are clearly limited in their capability of reasoning about 3D shape or viewpoints. In this work, we bring the worlds of 3D and 2D object representations closer, by building an object detector which leverages the expressive power of 3D object representations while at the same time can be robustly matched to image evidence. To that end, we gradually extend the successful deformable part model [1] to include viewpoint information and part-level 3D geometry information, resulting in several different models with different level of expressiveness. We end up with a 3D object model, consisting of multiple object parts represented in 3D and a continuous appearance model. We experimentally verify that our models, while providing richer object hypotheses than the 2D object models, provide consistently better joint object localization and viewpoint estimation than the state-of-the-art multi-view and 3D object detectors on various benchmarks (KITTI [2], 3D object classes [3], Pascal3D+ [4], Pascal VOC 2007 [5], EPFL multi-view cars [6]).

Current approaches to semantic image and scene understanding typically employ rather simple object representations such as 2D or 3D bounding boxes. While such coarse models are robust and allow for reliable object detection, they discard much of the information about objects' 3D shape and pose, and thus do not lend themselves well to higher-level reasoning. Here, we propose to base scene understanding on a high-resolution object representation. An object class--in our case cars--is modeled as a deformable 3D wireframe, which enables fine-grained modeling at the level of individual vertices and faces. We augment that model to explicitly include vertex-level occlusion, and embed all instances in a common coordinate frame, in order to infer and exploit object-object interactions. Specifically, from a single view we jointly estimate the shapes and poses of multiple objects in a common 3D frame. A ground plane in that frame is estimated by consensus among different objects, which significantly stabilizes monocular 3D pose estimation. The fine-grained model, in conjunction with the explicit 3D scene model, further allows one to infer part-level occlusions between the modeled objects, as well as occlusions by other, unmodeled scene elements. To demonstrate the benefits of such detailed object class models in the context of scene understanding we systematically evaluate our approach on the challenging KITTI street scene dataset. The experiments show that the model's ability to utilize image evidence at the level of individual parts improves monocular 3D pose estimation w.r.t. both location and (continuous) viewpoint.

We explore 3D layout estimation from a monocular image using detailed 3D object class models. We leverage the high geometric resolution provided by such models to reason about interactions among multiple object instances and demonstrate superior performance to using coarse 3D bounding box level hypotheses.

While the majority of today's object class models provide only 2D bounding boxes, far richer output hypotheses are desirable including viewpoint, fine-grained category, and 3D geometry estimate. However, models trained to provide richer output require larger amounts of training data, preferably well covering the relevant aspects such as viewpoint and fine-grained categories. In this paper, we address this issue from the perspective of transfer learning, and design an object class model that explicitly leverages correlations between visual features. Specifically, our model represents prior distributions over permissible multi-view detectors in a parametric way -- the priors are learned once from training data of a source object class, and can later be used to facilitate the learning of a detector for a target class. As we show in our experiments, this transfer is not only beneficial for detectors based on basic-level category representations, but also enables the robust learning of detectors that represent classes at finer levels of granularity, where training data is typically even scarcer and more unbalanced. As a result, we report largely improved performance in simultaneous 2D object localization and viewpoint estimation on a recent dataset of challenging street scenes.

We explore 3D scene reconstruction from a single view building on detailed 3D representations of object classes. These representations allow precise estimation of 3D shape and pose of individual objects, as well as explicit modeling of occlusions. We use the rich detail afforded by these models to reason about physical interactions among multiple object instances, and empirically demonstrate superior performance to models where no such reasoning is performed.

Current systems for scene understanding typically represent objects as 2D or 3D bounding boxes. While these representations have proven robust in a variety of applications, they provide only coarse approximations to the true 2D and 3D extent of objects. As a result, object-object interactions, such as occlusions or ground-plane contact, can be represented only superficially. In this paper, we approach the problem of scene understanding from the perspective of 3D shape modeling, and design a 3D scene representation that reasons jointly about the 3D shape of multiple objects. This representation allows to express 3D geometry and occlusion on the fine detail level of individual vertices of 3D wireframe models, and makes it possible to treat dependencies between objects, such as occlusion reasoning, in a deterministic way. In our experiments, we demonstrate the benefit of jointly estimating the 3D shape of multiple objects in a scene over working with coarse boxes, on the recently proposed KITTI dataset of realistic street scenes.

While 3D object representations are being revived in the context of multi-view object class detection and scene understanding, they have not yet attained wide-spread use in fine-grained categorization. State-of-the-art approaches achieve remarkable performance when training data is plentiful, but they are typically tied to flat, 2D representations that model objects as a collection of unconnected views, limiting their ability to generalize across viewpoints. In this paper, we therefore lift two state-of-the-art 2D object representations to 3D, on the level of both local feature appearance and location. In extensive experiments on existing and newly proposed datasets, we show our 3D object representations outperform their state-of-the-art 2D counterparts for fine-grained categorization and demonstrate their efficacy for estimating 3D geometry from images via ultra-wide baseline matching and 3D reconstruction.

In this work we introduce a large-scale, fine-grained dataset of cars. This dataset, consisting of 197 classes and 16,185 images, represents an order of magnitude increase in size over the only existing fine-grained car dataset [7] (14 classes, 1,904 images) and is comparable in size to the largest fine-grained datasets publicly available [9, 3]. The goals of this work are twofold: 1) to describe the difficulties encountered when collecting such a dataset and 2) to present baseline performance for two state-of-the-art methods.

Non-rigid object detection and articulated pose estimation are two related and challenging problems in computer vision. Numerous models have been proposed over the years and often address different special cases, such as pedestrian detection or upper body pose estimation in TV footage. This paper shows that such specialization may not be necessary, and proposes a generic approach based on the pictorial structures framework. We show that the right selection of components for both appearance and spatial modeling is crucial for general applicability and overall performance of the model. The appearance of body parts is modeled using densely sampled shape context descriptors and discriminatively rained AdaBoost classifiers. Furthermore, we interpret the normalized margin of each classifier as likelihood in a generative model. Non-Gaussian relationships between parts are represented as Gaussians in the coordinate system of the joint between parts. The marginal posterior of each part is inferred using belief propagation. We demonstrate that such a model is equally suitable for both detection and pose estimation tasks, outperforming the state of the art on three recently proposed d tasets.

Despite the success of recent object class recognition systems, the long-standing problem of partial occlusion remains a major challenge, and a principled solution is yet to be found. In this paper we leave the beaten path of methods that treat occlusion as just another source of noise - instead, we include the occluder itself into the modelling, by mining distinctive, reoccurring occlusion patterns from annotated training data. These patterns are then used as training data for dedicated detectors of varying sophistication. In particular, we evaluate and compare models that range from standard object class detectors to hierarchical, part-based representations of occluder/occludee pairs. In an extensive evaluation we derive insights that can aid further developments in tackling the occlusion challenge.

Despite the success of current state-of-the-art object class detectors, severe occlusion remains a major challenge. This is particularly true for more geometrically expressive 3D object class representations. While these representations have attracted renewed interest for precise object pose estimation, the focus has mostly been on rather clean datasets, where occlusion is not an issue. In this paper, we tackle the challenge of modeling occlusion in the context of a 3D geometric object class model that is capable of fine-grained, part-level 3D object reconstruction. Following the intuition that 3D modeling should facilitate occlusion reasoning, we design an explicit representation of likely geometric occlusion patterns. Robustness is achieved by pooling image evidence from of a set of fixed part detectors as well as a non-parametric representation of part configurations in the spirit of pose lets. We confirm the potential of our method on cars in a newly collected data set of inner-city street scenes with varying levels of occlusion, and demonstrate superior performance in occlusion estimation and part localization, compared to baselines that are unaware of occlusions.