Vision-language pre-training (VLP) on large-scale datasets has shown premier performance on various downstream tasks. A complete and fair benchmark (i.e., including large-scale pre-training datasets and diverse downstream tasks) is essential for VLP. While there are plenty of benchmarks with English corpus, building a rich benchmark for VLP with other languages, such as Chinese, remains a critical problem. To this end, we build a large-scale Chinese cross-modal benchmark called Zero for the research community to fairly compare VLP models. We release two pre-training datasets and five fine-tuning datasets for downstream tasks. Alongside, we propose a novel pre-training framework of pre-Ranking + Ranking for cross-modal learning. Specifically, we apply global contrastive pre-ranking to learn the individual representations of images and texts, respectively. We then fuse the representations in a fine-grained ranking manner via an image-text cross encoder and a text-image cross encoder. To further enhance the capability of the model, we propose a two-way distillation strategy consisting of target-guided Distillation and feature-guided Distillation. For brevity, we name our model R2D2. We achieve state-of-the-art performance on four public cross-modal datasets and the proposed five downstream datasets. When conducting zero-shot tasks on Flickr30k-CN, COCO-CN, and MUGE, R2D2 pre-trained on a 250 million dataset achieves significant improvements of 4.7%, 5.4%, and 6.3% in mean recall compared to the state-of-the-art. The datasets, models, and codes are available at //github.com/yuxie11/R2D2
相關內容
We summarize the results of a host of efforts using giant automatic speech recognition (ASR) models pre-trained using large, diverse unlabeled datasets containing approximately a million hours of audio. We find that the combination of pre-training, self-training and scaling up model size greatly increases data efficiency, even for extremely large tasks with tens of thousands of hours of labeled data. In particular, on an ASR task with 34k hours of labeled data, by fine-tuning an 8 billion parameter pre-trained Conformer model we can match state-of-the-art (SoTA) performance with only 3% of the training data and significantly improve SoTA with the full training set. We also report on the universal benefits gained from using big pre-trained and self-trained models for a large set of downstream tasks that cover a wide range of speech domains and span multiple orders of magnitudes of dataset sizes, including obtaining SoTA performance on many public benchmarks. In addition, we utilize the learned representation of pre-trained networks to achieve SoTA results on non-ASR tasks.
Recognizing scenes and objects in 3D from a single image is a longstanding goal of computer vision with applications in robotics and AR/VR. For 2D recognition, large datasets and scalable solutions have led to unprecedented advances. In 3D, existing benchmarks are small in size and approaches specialize in few object categories and specific domains, e.g. urban driving scenes. Motivated by the success of 2D recognition, we revisit the task of 3D object detection by introducing a large benchmark, called Omni3D. Omni3D re-purposes and combines existing datasets resulting in 234k images annotated with more than 3 million instances and 97 categories.3D detection at such scale is challenging due to variations in camera intrinsics and the rich diversity of scene and object types. We propose a model, called Cube R-CNN, designed to generalize across camera and scene types with a unified approach. We show that Cube R-CNN outperforms prior works on the larger Omni3D and existing benchmarks. Finally, we prove that Omni3D is a powerful dataset for 3D object recognition, show that it improves single-dataset performance and can accelerate learning on new smaller datasets via pre-training.
Recent breakthroughs in Natural Language Processing (NLP) have been driven by language models trained on a massive amount of plain text. While powerful, deriving supervision from textual resources is still an open question. For example, language model pretraining often neglects the rich, freely-available structures in textual data. In this thesis, we describe three lines of work that seek to improve the training and evaluation of neural models using naturally-occurring supervision. We first investigate self-supervised training losses to help enhance the performance of pretrained language models for various NLP tasks. Specifically, we alter the sentence prediction loss to make it better suited to other pretraining losses and more challenging to solve. We design an intermediate finetuning step that uses self-supervised training to promote models' ability in cross-task generalization. Then we describe methods to leverage the structures in Wikipedia and paraphrases. In particular, we propose training losses to exploit hyperlinks, article structures, and article category graphs for entity-, discourse-, entailment-related knowledge. We propose a framework that uses paraphrase pairs to disentangle semantics and syntax in sentence representations. We extend the framework for a novel generation task that controls the syntax of output text with a sentential exemplar. Lastly, we discuss our work on tailoring textual resources for establishing challenging evaluation tasks. We introduce three datasets by defining novel tasks using various fan-contributed websites, including a long-form data-to-text generation dataset, a screenplay summarization dataset, and a long-form story generation dataset. These datasets have unique characteristics offering challenges to future work in their respective task settings.
Video-Text Pre-training (VTP) aims to learn transferable representations for various downstream tasks from large-scale web videos. To date, almost all existing VTP methods are limited to retrieval-based downstream tasks, e.g., video retrieval, whereas their transfer potentials on localization-based tasks, e.g., temporal grounding, are under-explored. In this paper, we experimentally analyze and demonstrate the incompatibility of current VTP methods with localization tasks, and propose a novel Localization-oriented Video-Text Pre-training framework, dubbed as LocVTP. Specifically, we perform the fine-grained contrastive alignment as a complement to the coarse-grained one by a clip-word correspondence discovery scheme. To further enhance the temporal reasoning ability of the learned feature, we propose a context projection head and a temporal aware contrastive loss to perceive the contextual relationships. Extensive experiments on four downstream tasks across six datasets demonstrate that our LocVTP achieves state-of-the-art performance on both retrieval-based and localization-based tasks. Furthermore, we conduct comprehensive ablation studies and thorough analyses to explore the optimum model designs and training strategies.
Cognitive science has shown that humans perceive videos in terms of events separated by the state changes of dominant subjects. State changes trigger new events and are one of the most useful among the large amount of redundant information perceived. However, previous research focuses on the overall understanding of segments without evaluating the fine-grained status changes inside. In this paper, we introduce a new dataset called Kinetic-GEB+. The dataset consists of over 170k boundaries associated with captions describing status changes in the generic events in 12K videos. Upon this new dataset, we propose three tasks supporting the development of a more fine-grained, robust, and human-like understanding of videos through status changes. We evaluate many representative baselines in our dataset, where we also design a new TPD (Temporal-based Pairwise Difference) Modeling method for visual difference and achieve significant performance improvements. Besides, the results show there are still formidable challenges for current methods in the utilization of different granularities, representation of visual difference, and the accurate localization of status changes. Further analysis shows that our dataset can drive developing more powerful methods to understand status changes and thus improve video level comprehension. The dataset is available at //github.com/Yuxuan-W/GEB-Plus
Human motion trajectory prediction, an essential task for autonomous systems in many domains, has been on the rise in recent years. With a multitude of new methods proposed by different communities, the lack of standardized benchmarks and objective comparisons is increasingly becoming a major limitation to assess progress and guide further research. Existing benchmarks are limited in their scope and flexibility to conduct relevant experiments and to account for contextual cues of agents and environments. In this paper we present Atlas, a benchmark to systematically evaluate human motion trajectory prediction algorithms in a unified framework. Atlas offers data preprocessing functions, hyperparameter optimization, comes with popular datasets and has the flexibility to setup and conduct underexplored yet relevant experiments to analyze a method's accuracy and robustness. In an example application of Atlas, we compare five popular model- and learning-based predictors and find that, when properly applied, early physics-based approaches are still remarkably competitive. Such results confirm the necessity of benchmarks like Atlas.
Many basic indoor activities such as eating or writing are always conducted upon different tabletops (e.g., coffee tables, writing desks). It is indispensable to understanding tabletop scenes in 3D indoor scene parsing applications. Unfortunately, it is hard to meet this demand by directly deploying data-driven algorithms, since 3D tabletop scenes are rarely available in current datasets. To remedy this defect, we introduce TO-Scene, a large-scale dataset focusing on tabletop scenes, which contains 20,740 scenes with three variants. To acquire the data, we design an efficient and scalable framework, where a crowdsourcing UI is developed to transfer CAD objects from ModelNet and ShapeNet onto tables from ScanNet, then the output tabletop scenes are simulated into real scans and annotated automatically. Further, a tabletop-aware learning strategy is proposed for better perceiving the small-sized tabletop instances. Notably, we also provide a real scanned test set TO-Real to verify the practical value of TO-Scene. Experiments show that the algorithms trained on TO-Scene indeed work on the realistic test data, and our proposed tabletop-aware learning strategy greatly improves the state-of-the-art results on both 3D semantic segmentation and object detection tasks. Dataset and code are available at //github.com/GAP-LAB-CUHK-SZ/TO-Scene.
In many visual systems, visual tracking often bases on RGB image sequences, in which some targets are invalid in low-light conditions, and tracking performance is thus affected significantly. Introducing other modalities such as depth and infrared data is an effective way to handle imaging limitations of individual sources, but multi-modal imaging platforms usually require elaborate designs and cannot be applied in many real-world applications at present. Near-infrared (NIR) imaging becomes an essential part of many surveillance cameras, whose imaging is switchable between RGB and NIR based on the light intensity. These two modalities are heterogeneous with very different visual properties and thus bring big challenges for visual tracking. However, existing works have not studied this challenging problem. In this work, we address the cross-modal object tracking problem and contribute a new video dataset, including 654 cross-modal image sequences with over 481K frames in total, and the average video length is more than 735 frames. To promote the research and development of cross-modal object tracking, we propose a new algorithm, which learns the modality-aware target representation to mitigate the appearance gap between RGB and NIR modalities in the tracking process. It is plug-and-play and could thus be flexibly embedded into different tracking frameworks. Extensive experiments on the dataset are conducted, and we demonstrate the effectiveness of the proposed algorithm in two representative tracking frameworks against 17 state-of-the-art tracking methods. We will release the dataset for free academic usage, dataset download link and code will be released soon.
This paper presents Pix2Seq, a simple and generic framework for object detection. Unlike existing approaches that explicitly integrate prior knowledge about the task, we simply cast object detection as a language modeling task conditioned on the observed pixel inputs. Object descriptions (e.g., bounding boxes and class labels) are expressed as sequences of discrete tokens, and we train a neural net to perceive the image and generate the desired sequence. Our approach is based mainly on the intuition that if a neural net knows about where and what the objects are, we just need to teach it how to read them out. Beyond the use of task-specific data augmentations, our approach makes minimal assumptions about the task, yet it achieves competitive results on the challenging COCO dataset, compared to highly specialized and well optimized detection algorithms.
Since real-world objects and their interactions are often multi-modal and multi-typed, heterogeneous networks have been widely used as a more powerful, realistic, and generic superclass of traditional homogeneous networks (graphs). Meanwhile, representation learning (\aka~embedding) has recently been intensively studied and shown effective for various network mining and analytical tasks. In this work, we aim to provide a unified framework to deeply summarize and evaluate existing research on heterogeneous network embedding (HNE), which includes but goes beyond a normal survey. Since there has already been a broad body of HNE algorithms, as the first contribution of this work, we provide a generic paradigm for the systematic categorization and analysis over the merits of various existing HNE algorithms. Moreover, existing HNE algorithms, though mostly claimed generic, are often evaluated on different datasets. Understandable due to the application favor of HNE, such indirect comparisons largely hinder the proper attribution of improved task performance towards effective data preprocessing and novel technical design, especially considering the various ways possible to construct a heterogeneous network from real-world application data. Therefore, as the second contribution, we create four benchmark datasets with various properties regarding scale, structure, attribute/label availability, and \etc.~from different sources, towards handy and fair evaluations of HNE algorithms. As the third contribution, we carefully refactor and amend the implementations and create friendly interfaces for 13 popular HNE algorithms, and provide all-around comparisons among them over multiple tasks and experimental settings.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191