Wen Wang

Yuxin Fang, Wen Wang, Binhui Xie et al. · meta-ai

We launch EVA, a vision-centric foundation model to explore the limits of visual representation at scale using only publicly accessible data. EVA is a vanilla ViT pre-trained to reconstruct the masked out image-text aligned vision features conditioned on visible image patches. Via this pretext task, we can efficiently scale up EVA to one billion parameters, and sets new records on a broad range of representative vision downstream tasks, such as image recognition, video action recognition, object detection, instance segmentation and semantic segmentation without heavy supervised training. Moreover, we observe quantitative changes in scaling EVA result in qualitative changes in transfer learning performance that are not present in other models. For instance, EVA takes a great leap in the challenging large vocabulary instance segmentation task: our model achieves almost the same state-of-the-art performance on LVISv1.0 dataset with over a thousand categories and COCO dataset with only eighty categories. Beyond a pure vision encoder, EVA can also serve as a vision-centric, multi-modal pivot to connect images and text. We find initializing the vision tower of a giant CLIP from EVA can greatly stabilize the training and outperform the training from scratch counterpart with much fewer samples and less compute, providing a new direction for scaling up and accelerating the costly training of multi-modal foundation models. To facilitate future research, we release all the code and models at https://github.com/baaivision/EVA.

Xinlong Wang, Wen Wang, Yue Cao et al.

In-context learning, as a new paradigm in NLP, allows the model to rapidly adapt to various tasks with only a handful of prompts and examples. But in computer vision, the difficulties for in-context learning lie in that tasks vary significantly in the output representations, thus it is unclear how to define the general-purpose task prompts that the vision model can understand and transfer to out-of-domain tasks. In this work, we present Painter, a generalist model which addresses these obstacles with an "image"-centric solution, that is, to redefine the output of core vision tasks as images, and specify task prompts as also images. With this idea, our training process is extremely simple, which performs standard masked image modeling on the stitch of input and output image pairs. This makes the model capable of performing tasks conditioned on visible image patches. Thus, during inference, we can adopt a pair of input and output images from the same task as the input condition, to indicate which task to perform. Without bells and whistles, our generalist Painter can achieve competitive performance compared to well-established task-specific models, on seven representative vision tasks ranging from high-level visual understanding to low-level image processing. In addition, Painter significantly outperforms recent generalist models on several challenging tasks.

Xu Zhang, Wen Wang, Zhe Chen et al.

Animal pose estimation is challenging for existing image-based methods because of limited training data and large intra- and inter-species variances. Motivated by the progress of visual-language research, we propose that pre-trained language models (e.g., CLIP) can facilitate animal pose estimation by providing rich prior knowledge for describing animal keypoints in text. However, we found that building effective connections between pre-trained language models and visual animal keypoints is non-trivial since the gap between text-based descriptions and keypoint-based visual features about animal pose can be significant. To address this issue, we introduce a novel prompt-based Contrastive learning scheme for connecting Language and AniMal Pose (CLAMP) effectively. The CLAMP attempts to bridge the gap by adapting the text prompts to the animal keypoints during network training. The adaptation is decomposed into spatial-aware and feature-aware processes, and two novel contrastive losses are devised correspondingly. In practice, the CLAMP enables the first cross-modal animal pose estimation paradigm. Experimental results show that our method achieves state-of-the-art performance under the supervised, few-shot, and zero-shot settings, outperforming image-based methods by a large margin.

Chengzhi Jiang, Yanzhou Su, Wen Wang et al.

Existing classification-based face recognition methods have achieved remarkable progress, introducing large margin into hypersphere manifold to learn discriminative facial representations. However, the feature distribution is ignored. Poor feature distribution will wipe out the performance improvement brought about by margin scheme. Recent studies focus on the unbalanced inter-class distribution and form a equidistributed feature representations by penalizing the angle between identity and its nearest neighbor. But the problem is more than that, we also found the anisotropy of intra-class distribution. In this paper, we propose the `gradient-enhancing term' that concentrates on the distribution characteristics within the class. This method, named IntraLoss, explicitly performs gradient enhancement in the anisotropic region so that the intra-class distribution continues to shrink, resulting in isotropic and more compact intra-class distribution and further margin between identities. The experimental results on LFW, YTF and CFP-FP show that our outperforms state-of-the-art methods by gradient enhancement, demonstrating the superiority of our method. In addition, our method has intuitive geometric interpretation and can be easily combined with existing methods to solve the previously ignored problems.

Wen Wang, Honglei Zhuang, Mi Zhou et al.

Education has a significant impact on both society and personal life. With the development of technology, online education has been growing rapidly over the past decade. While there are several online education studies on student behavior analysis, the course concept mining, and course recommendations (Feng, Tang, and Liu 2019; Pan et al. 2017), there is little research on evaluating teachers' performance in online education. In this paper, we conduct a systematic study to understand and effectively predict teachers' performance using the subtitles of 1,085 online courses. Our model-free analysis shows that teachers' verbal cues (e.g., question strategy, emotional appealing, and hedging) and their course structure design are both significantly correlated with teachers' performance evaluation. Based on these insights, we then propose a hierarchical course BERT model to predict teachers' performance in online education. Our proposed model can capture the hierarchical structure within each course as well as the deep semantic features extracted from the course content. Experiment results show that our proposed method achieves significant gain over several state-of-the-art methods. Our study provides a significant social impact in helping teachers improve their teaching style and enhance their instructional material design for more effective online teaching in the future.

Wen Wang, Lijun Du, Yinxing Gao et al.

In this work, a discriminatively learned CNN embedding is proposed for remote sensing image scene classification. Our proposed siamese network simultaneously computes the classification loss function and the metric learning loss function of the two input images. Specifically, for the classification loss, we use the standard cross-entropy loss function to predict the classes of the images. For the metric learning loss, our siamese network learns to map the intra-class and inter-class input pairs to a feature space where intra-class inputs are close and inter-class inputs are separated by a margin. Concretely, for remote sensing image scene classification, we would like to map images from the same scene to feature vectors that are close, and map images from different scenes to feature vectors that are widely separated. Experiments are conducted on three different remote sensing image datasets to evaluate the effectiveness of our proposed approach. The results demonstrate that the proposed method achieves an excellent classification performance.

Haijun Liu, Jian Cheng, Wen Wang et al.

Deep metric learning aims at learning the distance metric between pair of samples, through the deep neural networks to extract the semantic feature embeddings where similar samples are close to each other while dissimilar samples are farther apart. A large amount of loss functions based on pair distances have been presented in the literature for guiding the training of deep metric learning. In this paper, we unify them in a general pair-based weighting loss function, where the minimizing objective loss is just the distances weighting of informative pairs. The general pair-based weighting loss includes two main aspects, (1) samples mining and (2) pairs weighting. Samples mining aims at selecting the informative positive and negative pair sets to exploit the structured relationship of samples in a mini-batch and also reduce the number of non-trivial pairs. Pair weighting aims at assigning different weights for different pairs according to the pair distances for discriminatively training the network. We detailedly review those existing pair-based losses inline with our general loss function, and explore some possible methods from the perspective of samples mining and pairs weighting. Finally, extensive experiments on three image retrieval datasets show that our general pair-based weighting loss obtains new state-of-the-art performance, demonstrating the effectiveness of the pair-based samples mining and pairs weighting for deep metric learning.

Katrin Kirchhoff, Bing Zhao, Wen Wang

Statistical machine translation for dialectal Arabic is characterized by a lack of data since data acquisition involves the transcription and translation of spoken language. In this study we develop techniques for extracting parallel data for one particular dialect of Arabic (Iraqi Arabic) from out-of-domain corpora in different dialects of Arabic or in Modern Standard Arabic. We compare two different data selection strategies (cross-entropy based and submodular selection) and demonstrate that a very small but highly targeted amount of found data can improve the performance of a baseline machine translation system. We furthermore report on preliminary experiments on using automatically translated speech data as additional training data.

Wen Wang, Zhen Cui, Hong Chang et al.

The comparison of heterogeneous samples extensively exists in many applications, especially in the task of image classification. In this paper, we propose a simple but effective coupled neural network, called Deeply Coupled Autoencoder Networks (DCAN), which seeks to build two deep neural networks, coupled with each other in every corresponding layers. In DCAN, each deep structure is developed via stacking multiple discriminative coupled auto-encoders, a denoising auto-encoder trained with maximum margin criterion consisting of intra-class compactness and inter-class penalty. This single layer component makes our model simultaneously preserve the local consistency and enhance its discriminative capability. With increasing number of layers, the coupled networks can gradually narrow the gap between the two views. Extensive experiments on cross-view image classification tasks demonstrate the superiority of our method over state-of-the-art methods.