2022-03-22 更新
Dual Contrastive Loss and Attention for GANs
Authors:Ning Yu, Guilin Liu, Aysegul Dundar, Andrew Tao, Bryan Catanzaro, Larry Davis, Mario Fritz
Generative Adversarial Networks (GANs) produce impressive results on unconditional image generation when powered with large-scale image datasets. Yet generated images are still easy to spot especially on datasets with high variance (e.g. bedroom, church). In this paper, we propose various improvements to further push the boundaries in image generation. Specifically, we propose a novel dual contrastive loss and show that, with this loss, discriminator learns more generalized and distinguishable representations to incentivize generation. In addition, we revisit attention and extensively experiment with different attention blocks in the generator. We find attention to be still an important module for successful image generation even though it was not used in the recent state-of-the-art models. Lastly, we study different attention architectures in the discriminator, and propose a reference attention mechanism. By combining the strengths of these remedies, we improve the compelling state-of-the-art Fr\’{e}chet Inception Distance (FID) by at least 17.5% on several benchmark datasets. We obtain even more significant improvements on compositional synthetic scenes (up to 47.5% in FID). Code and models are available at https://github.com/ningyu1991/AttentionDualContrastGAN .
PDF Accepted to ICCV’21
论文截图
Unsupervised Part Discovery from Contrastive Reconstruction
Authors:Subhabrata Choudhury, Iro Laina, Christian Rupprecht, Andrea Vedaldi
The goal of self-supervised visual representation learning is to learn strong, transferable image representations, with the majority of research focusing on object or scene level. On the other hand, representation learning at part level has received significantly less attention. In this paper, we propose an unsupervised approach to object part discovery and segmentation and make three contributions. First, we construct a proxy task through a set of objectives that encourages the model to learn a meaningful decomposition of the image into its parts. Secondly, prior work argues for reconstructing or clustering pre-computed features as a proxy to parts; we show empirically that this alone is unlikely to find meaningful parts; mainly because of their low resolution and the tendency of classification networks to spatially smear out information. We suggest that image reconstruction at the level of pixels can alleviate this problem, acting as a complementary cue. Lastly, we show that the standard evaluation based on keypoint regression does not correlate well with segmentation quality and thus introduce different metrics, NMI and ARI, that better characterize the decomposition of objects into parts. Our method yields semantic parts which are consistent across fine-grained but visually distinct categories, outperforming the state of the art on three benchmark datasets. Code is available at the project page: https://www.robots.ox.ac.uk/~vgg/research/unsup-parts/.
PDF NeurIPS 2021. Project page: https://www.robots.ox.ac.uk/~vgg/research/unsup-parts/
论文截图
Partitioning Image Representation in Contrastive Learning
Authors:Hyunsub Lee, Heeyoul Choi
In contrastive learning in the image domain, the anchor and positive samples are forced to have as close representations as possible. However, forcing the two samples to have the same representation could be misleading because the data augmentation techniques make the two samples different. In this paper, we introduce a new representation, partitioned representation, which can learn both common and unique features of the anchor and positive samples in contrastive learning. The partitioned representation consists of two parts: the content part and the style part. The content part represents common features of the class, and the style part represents the own features of each sample, which can lead to the representation of the data augmentation method. We can achieve the partitioned representation simply by decomposing a loss function of contrastive learning into two terms on the two separate representations, respectively. To evaluate our representation with two parts, we take two framework models: Variational AutoEncoder (VAE) and BootstrapYour Own Latent(BYOL) to show the separability of content and style, and to confirm the generalization ability in classification, respectively. Based on the experiments, we show that our approach can separate two types of information in the VAE framework and outperforms the conventional BYOL in linear separability and a few-shot learning task as downstream tasks.
PDF 6 pages, 8 figures, 2 tables
论文截图
Regional Semantic Contrast and Aggregation for Weakly Supervised Semantic Segmentation
Authors:Tianfei Zhou, Meijie Zhang, Fang Zhao, Jianwu Li
Learning semantic segmentation from weakly-labeled (e.g., image tags only) data is challenging since it is hard to infer dense object regions from sparse semantic tags. Despite being broadly studied, most current efforts directly learn from limited semantic annotations carried by individual image or image pairs, and struggle to obtain integral localization maps. Our work alleviates this from a novel perspective, by exploring rich semantic contexts synergistically among abundant weakly-labeled training data for network learning and inference. In particular, we propose regional semantic contrast and aggregation (RCA) . RCA is equipped with a regional memory bank to store massive, diverse object patterns appearing in training data, which acts as strong support for exploration of dataset-level semantic structure. Particularly, we propose i) semantic contrast to drive network learning by contrasting massive categorical object regions, leading to a more holistic object pattern understanding, and ii) semantic aggregation to gather diverse relational contexts in the memory to enrich semantic representations. In this manner, RCA earns a strong capability of fine-grained semantic understanding, and eventually establishes new state-of-the-art results on two popular benchmarks, i.e., PASCAL VOC 2012 and COCO 2014.
PDF Accepted to CVPR 2022. Code: https://github.com/maeve07/RCA.git
论文截图
SimCVD: Simple Contrastive Voxel-Wise Representation Distillation for Semi-Supervised Medical Image Segmentation
Authors:Chenyu You, Yuan Zhou, Ruihan Zhao, Lawrence Staib, James S. Duncan
Automated segmentation in medical image analysis is a challenging task that requires a large amount of manually labeled data. However, most existing learning-based approaches usually suffer from limited manually annotated medical data, which poses a major practical problem for accurate and robust medical image segmentation. In addition, most existing semi-supervised approaches are usually not robust compared with the supervised counterparts, and also lack explicit modeling of geometric structure and semantic information, both of which limit the segmentation accuracy. In this work, we present SimCVD, a simple contrastive distillation framework that significantly advances state-of-the-art voxel-wise representation learning. We first describe an unsupervised training strategy, which takes two views of an input volume and predicts their signed distance maps of object boundaries in a contrastive objective, with only two independent dropout as mask. This simple approach works surprisingly well, performing on the same level as previous fully supervised methods with much less labeled data. We hypothesize that dropout can be viewed as a minimal form of data augmentation and makes the network robust to representation collapse. Then, we propose to perform structural distillation by distilling pair-wise similarities. We evaluate SimCVD on two popular datasets: the Left Atrial Segmentation Challenge (LA) and the NIH pancreas CT dataset. The results on the LA dataset demonstrate that, in two types of labeled ratios (i.e., 20% and 10%), SimCVD achieves an average Dice score of 90.85% and 89.03% respectively, a 0.91% and 2.22% improvement compared to previous best results. Our method can be trained in an end-to-end fashion, showing the promise of utilizing SimCVD as a general framework for downstream tasks, such as medical image synthesis, enhancement, and registration.
PDF IEEE Transactions on Medical Imaging (IEEE-TMI) 2022
论文截图
M5Product: Self-harmonized Contrastive Learning for E-commercial Multi-modal Pretraining
Authors:Xiao Dong, Xunlin Zhan, Yangxin Wu, Yunchao Wei, Michael C. Kampffmeyer, Xiaoyong Wei, Minlong Lu, Yaowei Wang, Xiaodan Liang
Despite the potential of multi-modal pre-training to learn highly discriminative feature representations from complementary data modalities, current progress is being slowed by the lack of large-scale modality-diverse datasets. By leveraging the natural suitability of E-commerce, where different modalities capture complementary semantic information, we contribute a large-scale multi-modal pre-training dataset M5Product. The dataset comprises 5 modalities (image, text, table, video, and audio), covers over 6,000 categories and 5,000 attributes, and is 500 larger than the largest publicly available dataset with a similar number of modalities. Furthermore, M5Product contains incomplete modality pairs and noise while also having a long-tailed distribution, resembling most real-world problems. We further propose Self-harmonized ContrAstive LEarning (SCALE), a novel pretraining framework that integrates the different modalities into a unified model through an adaptive feature fusion mechanism, where the importance of each modality is learned directly from the modality embeddings and impacts the inter-modality contrastive learning and masked tasks within a multi-modal transformer model. We evaluate the current multi-modal pre-training state-of-the-art approaches and benchmark their ability to learn from unlabeled data when faced with the large number of modalities in the M5Product dataset. We conduct extensive experiments on four downstream tasks and demonstrate the superiority of our SCALE model, providing insights into the importance of dataset scale and diversity.
PDF CVPR2022
论文截图