Glance-and-Gaze Vision Transformer

Recently, there emerges a series of vision Transformers, which show superior performance with a more compact model size than conventional convolutional neural networks, thanks to the strong ability of Transformers to model long-range dependencies. However, the advantages of vision Transformers also come with a price: Self-attention, the core part of Transformer, has a quadratic complexity to the input sequence length. This leads to a dramatic increase of computation and memory cost with the increase of sequence length, thus introducing difficulties when applying Transformers to the vision tasks that require dense predictions based on high-resolution feature maps. In this paper, we propose a new vision Transformer, named Glance-and-Gaze Transformer (GG-Transformer), to address the aforementioned issues. It is motivated by the Glance and Gaze behavior of human beings when recognizing objects in natural scenes, with the ability to efficiently model both long-range dependencies and local context. In GG-Transformer, the Glance and Gaze behavior is realized by two parallel branches: The Glance branch is achieved by performing self-attention on the adaptively-dilated partitions of the input, which leads to a linear complexity while still enjoying a global receptive field; The Gaze branch is implemented by a simple depth-wise convolutional layer, which compensates local image context to the features obtained by the Glance mechanism. We empirically demonstrate our method achieves consistently superior performance over previous state-of-the-art Transformers on various vision tasks and benchmarks. The codes and models will be made available at https://github.com/yucornetto/GG-Transformer.

Unsupervised Part Discovery via Feature Alignment

Understanding objects in terms of their individual parts is important, because it enables a precise understanding of the objects geometrical structure, and enhances object recognition when the object is seen in a novel pose or under partial occlusion. However, the manual annotation of parts in large scale datasets is time consuming and expensive. In this paper, we aim at discovering object parts in an unsupervised manner, i.e., without ground-truth part or keypoint annotations. Our approach builds on the intuition that objects of the same class in a similar pose should have their parts aligned at similar spatial locations. We exploit the property that neural network features are largely invariant to nuisance variables and the main remaining source of variations between images of the same object category is the object pose. Specifically, given a training image, we find a set of similar images that show instances of the same object category in the same pose, through an affine alignment of their corresponding feature maps. The average of the aligned feature maps serves as pseudo ground-truth annotation for a supervised training of the deep network backbone. During inference, part detection is simple and fast, without any extra modules or overheads other than a feed-forward neural network. Our experiments on several datasets from different domains verify the effectiveness of the proposed method. For example, we achieve 37.8 mAP on VehiclePart, which is at least 4.2 better than previous methods.

Can Temporal Information Help with Contrastive Self-Supervised Learning?

Leveraging temporal information has been regarded as essential for developing video understanding models. However, how to properly incorporate temporal information into the recent successful instance discrimination based contrastive self-supervised learning (CSL) framework remains unclear. As an intuitive solution, we find that directly applying temporal augmentations does not help, or even impair video CSL in general. This counter-intuitive observation motivates us to re-design existing video CSL frameworks, for better integration of temporal knowledge. To this end, we present Temporal-aware Contrastive self-supervised learningTaCo, as a general paradigm to enhance video CSL. Specifically, TaCo selects a set of temporal transformations not only as strong data augmentation but also to constitute extra self-supervision for video understanding. By jointly contrasting instances with enriched temporal transformations and learning these transformations as self-supervised signals, TaCo can significantly enhance unsupervised video representation learning. For instance, TaCo demonstrates consistent improvement in downstream classification tasks over a list of backbones and CSL approaches. Our best model achieves 85.1% (UCF-101) and 51.6% (HMDB-51) top-1 accuracy, which is a 3% and 2.4% relative improvement over the previous state-of-the-art.

CoKe: Localized Contrastive Learning for Robust Keypoint Detection

Todays most popular approaches to keypoint detection involve very complex network architectures that aim to learn holistic representations of all keypoints. In this work, we take a step back and ask: Can we simply learn a local keypoint representation from the output of a standard backbone architecture? This will help make the network simpler and more robust, particularly if large parts of the object are occluded. We demonstrate that this is possible by looking at the problem from the perspective of representation learning. Specifically, the keypoint kernels need to be chosen to optimize three types of distances in the feature space: Features of the same keypoint should be similar to each other, while differing from those of other keypoints, and also being distinct from features from the background clutter. We formulate this optimization process within a framework, which we call CoKe, which includes supervised contrastive learning. CoKe needs to make several approximations to enable representation learning process on large datasets. In particular, we introduce a clutter bank to approximate non-keypoint features, and a momentum update to compute the keypoint representation while training the feature extractor. Our experiments show that CoKe achieves state-of-the-art results compared to approaches that jointly represent all keypoints holistically (Stacked Hourglass Networks, MSS-Net) as well as to approaches that are supervised by detailed 3D object geometry (StarMap). Moreover, CoKe is robust and performs exceptionally well when objects are partially occluded and significantly outperforms related work on a range of diverse datasets (PASCAL3D+, MPII, ObjectNet3D).

CLEVR-Ref+: Diagnosing Visual Reasoning with Referring Expressions

Referring object detection and referring image segmentation are important tasks that require joint understanding of visual information and natural language. Yet there has been evidence that current benchmark datasets suffer from bias, and current state-of-the-art models cannot be easily evaluated on their intermediate reasoning process. To address these issues and complement similar efforts in visual question answering, we build CLEVR-Ref+, a synthetic diagnostic dataset for referring expression comprehension. The precise locations and attributes of the objects are readily available, and the referring expressions are automatically associated with functional programs. The synthetic nature allows control over dataset bias (through sampling strategy), and the modular programs enable intermediate reasoning ground truth without human annotators. In addition to evaluating several state-of-the-art models on CLEVR-Ref+, we also propose IEP-Ref, a module network approach that significantly outperforms other models on our dataset. In particular, we present two interesting and important findings using IEP-Ref: (1) the module trained to transform feature maps into segmentation masks can be attached to any intermediate module to reveal the entire reasoning process step-by-step; (2) even if all training data has at least one object referred, IEP-Ref can correctly predict no-foreground when presented with false-premise referring expressions. To the best of our knowledge, this is the first direct and quantitative proof that neural modules behave in the way they are intended.

Semantic Part Detection via Matching: Learning to Generalize to Novel Viewpoints from Limited Training Data

Detecting semantic parts of an object is a challenging task in computer vision, particularly because it is hard to construct large annotated datasets due to the difficulty of annotating semantic parts. In this paper we present an approach which learns from a small training dataset of annotated semantic parts, where the object is seen from a limited range of viewpoints, but generalizes to detect semantic parts from a much larger range of viewpoints. Our approach is based on a matching algorithm for finding accurate spatial correspondence between two images, which enables semantic parts annotated on one image to be transplanted to another. In particular, this enables images in the training dataset to be matched to a virtual 3D model of the object (for simplicity, we assume that the object viewpoint can be estimated by standard techniques). Then a clustering algorithm is used to annotate the semantic parts of the 3D virtual model. This virtual 3D model can be used to synthesize annotated images from a large range of viewpoint. These can be matched to images in the test set, using the same matching algorithm, to detect semantic parts in novel viewpoints of the object. Our algorithm is very simple, intuitive, and contains very few parameters. We evaluate our approach in the car subclass of the VehicleSemanticPart dataset. We show it outperforms standard deep network approaches and, in particular, performs much better on novel viewpoints. For facilitating the future research, code is available: https://github.com/ytongbai/SemanticPartDetection