ECCV 2022 Papers

"Transformers and masked language modeling are quickly being adopted and explored in computer vision as vision transformers and masked image modeling (MIM). In this work, we argue that image token masking differs from token masking in text, due to the amount and correlation of tokens in an image. In particular, to generate a challenging pretext task for MIM, we advocate a shift from random masking to informed masking. We develop and exhibit this idea in the context of distillation-based MIM, where a teacher transformer encoder generates an attention map, which we use to guide masking for the student. We thus introduce a novel masking strategy, called attention-guided masking (AttMask), and we demonstrate its effectiveness over random masking for dense distillation-based MIM as well as plain distillation-based self-supervised learning on classification tokens. We confirm that AttMask accelerates the learning process and improves the performance on a variety of downstream tasks. We provide the implementation code at https://github.com/gkakogeorgiou/attmask."

"3D scene flow estimation from point clouds is a low-level 3D motion perception task in computer vision. Flow embedding is a commonly used technique in scene flow estimation, and it encodes the point motion between two consecutive frames. Thus, it is critical for the flow embeddings to capture the correct overall direction of the motion. However, previous works only search locally to determine a soft correspondence, ignoring the distant points that turn out to be the actual matching ones. In addition, the estimated correspondence is usually from the forward direction of the adjacent point clouds, and may not be consistent with the estimated correspondence acquired from the backward direction. To tackle these problems, we propose a novel all-to-all flow embedding layer with backward reliability validation during the initial scene flow estimation. Besides, we investigate and compare several design choices in key components of the 3D scene flow network, including the point similarity calculation, input elements of predictor, and predictor & refinement level design. After carefully choosing the most effective designs, we are able to present a model that achieves the state-of-the-art performance on FlyingThings3D and KITTI Scene Flow datasets. Our proposed model surpasses all existing methods by at least 38.2% on FlyingThings3D dataset and 24.7% on KITTI Scene Flow dataset for EPE3D metric. We release our codes at https://github.com/IRMVLab/3DFlow."

"Modern iterations of deep learning models contain millions (billions) of unique parameters--each represented by a $b$-bit number. Popular attempts at compressing neural networks (such as pruning and quantisation) have shown that many of the parameters are superfluous, which we can remove (pruning) or express with $b’ < b$ bits (quantisation) without hindering performance. Here we look to go much further in minimising the information content of networks. Rather than a channel or layer-wise encoding, we look to lossless whole-network quantisation to minimise the entropy and number of unique parameters in a network. We propose a new method, which we call Weight Fixing Networks (WFN) that we design to realise four model outcome objectives: i) very few unique weights, ii) low-entropy weight encodings, iii) unique weight values which are amenable to energy-saving versions of hardware multiplication, and iv) lossless task-performance. Some of these goals are conflicting. To best balance these conflicts, we combine a few novel (and some well-trodden) tricks; a novel regularisation term, (i, ii) a view of clustering cost as relative distance change (i, ii, iv), and a focus on whole-network re-use of weights (i, iii). Our Imagenet experiments demonstrate lossless compression using 56x fewer unique weights and a 1.9x lower weight-space entropy than SOTA quantisation approaches."

"General purpose vision (GPV) systems are models that are designed to solve a wide array of visual tasks without requiring architectural changes. Today, GPVs primarily learn both skills and concepts from large fully supervised datasets. Scaling GPVs to tens of thousands of concepts by acquiring data to learn each concept for every skill quickly becomes prohibitive. This work presents an effective and inexpensive alternative: learn skills from supervised datasets, learn concepts from web image search, and leverage a key characteristic of GPVs -- the ability to transfer visual knowledge across skills. We use a dataset of 1M+ images spanning 10k+ visual concepts to demonstrate webly-supervised concept expansion for two existing GPVs (GPV-1 and VL-T5) on 3 benchmarks - 5 COCO based datasets (80 primary concepts), a newly curated series of 5 datasets based on the OpenImages and VisualGenome repositories ( 500 concepts) and the Web-derived dataset (10k+ concepts). We also propose a new architecture, GPV-2 that supports a variety of tasks -- from vision tasks like classification and localization to vision+language tasks like QA and captioning to more niche ones like human-object interaction detection. GPV-2 benefits hugely from web data and outperforms GPV-1 and VL-T5 across these benchmarks."

"Action understanding has evolved into the era of fine granularity, as most human behaviors in real life have only minor differences. To detect these fine-grained actions accurately in a label-efficient way, we tackle the problem of weakly-supervised fine-grained temporal action detection in videos for the first time. Without the careful design to capture subtle differences between fine-grained actions, previous weakly-supervised models for general action detection cannot perform well in fine-grained settings. We propose to model actions as the combinations of reusable atomic actions which are automatically discovered from data through self-supervised clustering, in order to capture the commonality and individuality of fine-grained actions. The learnt atomic actions, represented by visual concepts, are further mapped to fine and coarse action labels leveraging the semantic label hierarchy. Our approach constructs a visual representation hierarchy of four levels: clip level, atomic action level, fine action class level and coarse action class level, with supervision at each level. Extensive experiments on two large-scale fine-grained video datasets, FineAction and FineGym, show the benefit of our proposed weakly-supervised model for fine-grained action detection, and it achieves state-of-the-art results."

"Recently, there has been growing attention on an end-to-end deep learning-based stitching model. However, the most challenging point in deep learning-based stitching is to obtain pairs of input images with a narrow field of view and ground truth images with a wide field of view captured from real-world scenes. To overcome this difficulty, we develop a weakly-supervised learning mechanism to train the stitching model without requiring genuine ground truth images. In addition, we propose a stitching model that takes multiple real-world fisheye images as inputs and creates a 360$^{\circ}$ output image in an equirectangular projection format. In particular, our model consists of color consistency corrections, warping, and blending, and is trained by perceptual and SSIM losses. The effectiveness of the proposed algorithm is verified on two real-world stitching datasets."

"Weakly Supervised Object Localization (WSOL), which aims to localize objects by only using image-level labels, has attracted much attention because of its low annotation cost in real applications. Recent studies leverage the advantage of self-attention in visual Transformer for long-range dependency to re-active semantic regions, aiming to avoid partial activation in traditional class activation mapping (CAM). However, the long-range modeling in Transformer neglects the inherent spatial coherence of the object, and it usually diffuses the semantic-aware regions far from the object boundary, making localization results significantly larger or far smaller. To address such an issue, we introduce a simple yet effective Spatial Calibration Module (SCM) for accurate WSOL, incorporating semantic similarities of patch tokens and their spatial relationships into a unified diffusion model. Specifically, we introduce a learnable parameter to dynamically adjust the semantic correlations and spatial context intensities for effective information propagation. In practice, SCM is designed as an external module of Transformer, and can be removed during inference to reduce the computation cost. The object-sensitive localization ability is implicitly embedded into the Transformer encoder through optimization in the training phase. It enables the generated attention maps to capture the sharper object boundaries and filter the object-irrelevant background area. Extensive experimental results demonstrate the effectiveness of the proposed method, which significantly outperforms its counterpart TS-CAM on both CUB-200 and ImageNet-1K benchmarks. The code is available at https://github.com/164140757/SCM."

"Weakly supervised object localization (WSOL) aims at detecting objects through only image-level labels. Class activation maps (CAMs) are the commonly used features for WSOL. However, existing CAM-based methods tend to excessively pursue discriminative features for object recognition and hence ignore the feature similarities among different categories, thereby leading to CAMs incomplete for object localization. In addition, CAMs are sensitive to background noise due to over-dependence on the holistic classification. In this paper, we propose a simple but effective WSOL model (named ISIC) through Inter-class feature Similarity and Intra-class appearance Consistency. In practice, our ISIC model first proposes the inter-class feature similarity (ICFS) loss against the original cross entropy loss. Such an ICFS loss sufficiently leverages the shared features together with the discriminative features between different categories, which significantly reduces the model over-fitting risk to background noise and brings more complete object masks. Besides, instead of CAMs, a non-negative matrix factorization mask module is applied to extract object masks from multiple intra-class images. Thanks to intra-class appearance consistency, the achieved pseudo masks are more complete and robust. As a result, extensive experiments confirm that our ISIC model achieves state-of-the-art on both CUB-200 and ImageNet-1K benchmarks i.e., 97.3% and 70.0% GT-Known localization accuracy, respectively."

"Transformers for visual-language representation learning have been getting a lot of interest and shown tremendous performance on visual question answering (VQA) and grounding. However, most systems that show good performance of those tasks still rely on pre-trained object detectors during training, which limits their applicability to the object classes available for those detectors. To mitigate this limitation, this paper focuses on the problem of weakly supervised grounding in the context of visual question answering in transformers. Our approach leverages capsules by transforming each visual token into a capsule representation in the visual encoder; it then uses activations from language self-attention layers as a text-guided selection module to mask those capsules before they are forwarded to the next layer. We evaluate our approach on the challenging GQA as well as VQA-HAT dataset for VQA grounding. Our experiments show that: while removing the information of masked objects from standard transformer architectures leads to a significant drop in performance, the integration of capsules significantly improves the grounding ability of such systems and provides new state-of-the-art results compared to other approaches in the field."

"Current state-of-the-art 3D scene understanding methods are merely designed in a full-supervised way. However, in the limited reconstruction cases, only limited 3D scenes can be reconstructed and annotated. We are in need of a framework that can concurrently be applied to 3D point cloud semantic segmentation and instance segmentation, particularly in circumstances where labels are rather scarce. The paper introduces an effective approach to tackle the 3D scene understanding problem when labeled scenes are limited. To leverage the boundary information, we propose a novel energy-based loss with boundary awareness benefiting from the region-level boundary labels predicted by the boundary prediction network. To encourage latent instance discrimination and guarantee efficiency, we propose the first unsupervised region-level semantic contrastive learning scheme for point clouds, which uses confident predictions of the network to discriminate the intermediate feature embeddings in multiple stages. In the limited reconstruction case, our proposed approach, termed WS3D, has pioneer performance on the large-scale ScanNet on semantic segmentation and instance segmentation. Also, our proposed WS3D achieves state-of-the-art performance on the other indoor and outdoor datasets S3DIS and SemanticKITTI."

"Object pose estimation is a crucial task in computer vision and augmented reality. One of its key challenges is the difficulty of annotation of real training data and the lack of textured CAD models. Therefore, pipelines which do not require CAD models and which can be trained with few labeled images are desirable. We propose a weakly-supervised approach for object pose estimation from RGB-D data using training sets composed of very few labeled images with pose annotations along with weakly-labeled images with ground truth segmentation masks without pose labels. We achieve this by learning to annotate weakly-labeled training data through shape alignment while simultaneously training a pose prediction network. Point cloud alignment is performed using structure and rotation-invariant feature-based losses. We further learn an implicit shape representation, which allows the method to work without the known CAD model and also contributes to pose alignment and pose refinement during training on weakly labeled images. The experimental evaluation shows that our method achieves state-of-the-art results on LineMOD, Occlusion-LineMOD and TLess despite being trained using relative poses and on only a fraction of labeled data used by the other methods. We also achieve comparable results to state-of-the-art RGB-D based pose estimation approaches even when further reducing the amount of unlabeled training data. In addition, our method works even if relative camera poses are given instead of object pose annotations which are typically easier to obtain."

"Panoptic image segmentation is the computer vision task of finding groups of pixels in an image and assigning semantic classes and object instance identifiers to them. Research in image segmentation has become increasingly popular due to its critical applications in robotics and autonomous driving. The research community thereby relies on publicly available benchmark dataset to advance the state-of-the-art in computer vision. Due to the high costs of densely labeling the images, however, there is a shortage of publicly available ground truth labels that are suitable for panoptic segmentation. The high labeling costs also make it challenging to extend existing datasets to the video domain and to multi-camera setups. We therefore present the Waymo Open Dataset: Panoramic Video Panoptic Segmentation dataset, a large-scale dataset that offers high-quality panoptic segmentation labels for autonomous driving. We generate our dataset using the publicly available Waymo Open Dataset (WOD), leveraging the diverse set of camera images. Our labels are consistent over time for video processing and consistent across multiple cameras mounted on the vehicles for full panoramic scene understanding. Specifically, we offer labels for 28 semantic categories and 2,860 temporal sequences that were captured by five cameras mounted on autonomous vehicles driving in three different geographical locations, leading to a total of 100k labeled camera images. To the best of our knowledge, this makes our dataset an order of magnitude larger than existing datasets that offer video panoptic segmentation labels. We further propose a new benchmark for Panoramic Video Panoptic Segmentation and establish a number of strong baselines based on the DeepLab family of models. We have made the benchmark and the code publicly available, which we hope will facilitate future research on holistic scene understanding. Our dataset can be found at: https://waymo.com/open/"

"Existing few-shot image generation approaches typically employ fusion-based strategies, either on the image or the feature level, to produce new images. However, previous approaches struggle to synthesize high-frequency signals with fine details, deteriorating the synthesis quality. To address this, we propose WaveGAN, a frequency-aware model for few-shot image generation. Concretely, we disentangle encoded features into multiple frequency components and perform low-frequency skip connections to preserve outline and structural information. Then we alleviate the generator’s struggles of synthesizing fine details by employing high-frequency skip connections, thus providing informative frequency information to the generator. Moreover, we utilize a frequency L1-loss on the generated and real images to further impede frequency information loss. Extensive experiments demonstrate the effectiveness and advancement of our method on three datasets. Noticeably, we achieve new state-of-the-art with FID 42.17, LPIPS 0.3868, FID 30.35, LPIPS 0.5076, and FID 4.96, LPIPS 0.3822 respectively on Flower, Animal Faces, and VGGFace. GitHub: https://github.com/kobeshegu/ECCV2022_WaveGAN"

"Multi-scale Vision Transformer (ViT) has emerged as a powerful backbone for computer vision tasks, while the self-attention computation in Transformer scales quadratically w.r.t. the input patch number. Thus, existing solutions commonly employ down-sampling operations (e.g., average pooling) over keys/values to dramatically reduce the computational cost. In this work, we argue that such over-aggressive down-sampling design is not invertible and inevitably causes information dropping especially for high-frequency components in objects (e.g., texture details). Motivated by the wavelet theory, we construct a new Wavelet Vision Transformer (Wave-ViT) that formulates the invertible down-sampling with wavelet transforms and self-attention learning in a unified way. This proposal enables self-attention learning with lossless down-sampling over keys/values, facilitating the pursuing of a better efficiency-vs-accuracy trade-off. Furthermore, inverse wavelet transforms are leveraged to strengthen self-attention outputs by aggregating local contexts with enlarged receptive field. We validate the superiority of Wave-ViT through extensive experiments over multiple vision tasks (e.g., image recognition, object detection and instance segmentation). Its performances surpass state-of-the-art ViT backbones with comparable FLOPs. Source code is available at \url{https://github.com/YehLi/ImageNetModel}."

"As a common security tool, visible watermarking has been widely applied to protect copyrights of digital images. However, recent works have shown that visible watermarks can be removed by DNNs without damaging their host images. Such watermark-removal techniques pose a great threat to the ownership of images. Inspired by the vulnerability of DNNs on adversarial perturbations, we propose a novel defence mechanism by adversarial machine learning for good. From the perspective of the adversary, blind watermark-removal networks can be posed as our target models; then we actually optimize an imperceptible adversarial perturbation on the host images to proactively attack against watermark-removal networks, dubbed Watermark Vaccine. Specifically, two types of vaccines are proposed. Disrupting Watermark Vaccine (DWV) induces to ruin the host image along with watermark after passing through watermark-removal networks. In contrast, Inerasable Watermark Vaccine (IWV) works in another fashion of trying to keep the watermark not removed and still noticeable. Extensive experiments demonstrate the effectiveness of our DWV/IWV in preventing watermark removal, especially on various watermark removal networks. The Code is released in https://github.com/thinwayliu/Watermark-Vaccine."

"Image-based artistic rendering can synthesize a variety of expressive styles using algorithmic image filtering. In contrast to deep learning-based methods, these heuristics-based filtering techniques can operate on high-resolution images, are interpretable, and can be parameterized according to various design aspects. However, adapting or extending these techniques to produce new styles is often a tedious and error-prone task that requires expert knowledge. We propose a new paradigm to alleviate this problem: implementing algorithmic image filtering techniques as differentiable operations that can learn parametrizations aligned to certain reference styles. To this end, we present WISE, an example-based image-processing system that can handle a multitude of stylization techniques, such as watercolor, oil, or cartoon stylization, within a common framework. By training parameter prediction networks for global and local filter parameterizations, we can simultaneously adapt effects to reference styles and image content, e.g., to enhance facial features. Our method can be optimized in a style-transfer framework or learned in a generative-adversarial setting for image-to-image translation. We demonstrate that jointly training an xDoG filter and a CNN for postprocessing can achieve comparable results to a state-of-the-art GAN-based method."

"Weakly-supervised object detection (WSOD) aims to train an object detector only requiring the image-level annotations. Recently, some works have managed to select the accurate boxes generated from a well-trained WSOD network to supervise a semi-supervised detection framework for better performance. However, these approaches simply divide the training set into labeled and unlabeled sets according to the image-level criteria, such that sufficient mislabeled or wrongly localized box predictions are chosen as pseudo ground-truths, resulting in a sub-optimal solution of detection performance. To overcome this issue, we propose a novel WSOD framework with a new paradigm that switches from weak supervision to noisy supervision (W2N).Generally, with given pseudo ground-truths generated from the well-trained WSOD network, we propose a two-module iterative training algorithm to refine pseudo labels and supervise better object detector progressively. In the localization adaptation module, we propose a regularization loss to reduce the proportion of discriminative parts in original pseudo ground-truths, obtaining better pseudo ground-truths for further training. In the semi-supervised module, we propose a two tasks instance-level split method to select high-quality labels for training a semi-supervised detector. Experimental results on different benchmarks verify the effectiveness of W2N, and our W2N outperforms all existing pure WSOD methods and transfer learning methods."

"We propose a novel keypoint voting scheme based on intersecting spheres, that is more accurate than existing schemes and allows for fewer, more disperse keypoints. The scheme is based upon the distance between points, which as a 1D quantity can be regressed more accurately than the 2D and 3D vector and offset quantities regressed in previous work, yielding more accurate keypoint localization. The scheme forms the basis of the proposed RCVPose method for 6 DoF pose estimation of 3D objects in RGB-D data, which is particularly effective at handling occlusions. A CNN is trained to estimate the distance between the 3D point corresponding to the depth mode of each RGB pixel, and a set of 3 disperse keypoints defined in the object frame. At inference, a sphere centered at each 3D point is generated, of radius equal to this estimated distance. The surfaces of these spheres vote to increment a 3D accumulator space, the peaks of which indicate keypoint locations. The proposed radial voting scheme is more accurate than previous vector or offset schemes, and is robust to disperse keypoints. Experiments demonstrate RCVPose to be highly accurate and competitive, achieving state-of-theart results on the LINEMOD (99.7%) and YCB-Video (97.2%) datasets, notably scoring +4.9% higher (71.1%) than previous methods on the challenging Occlusion LINEMOD dataset, and on average outperforming all other published results from the BOP benchmark for these 3 datasets. Our code is available at http://www.github.com/aaronwool/rcvpose."

"This paper presents an audio-visual approach for voice separation which produces state-of-the-art results at a low latency in two scenarios: speech and singing voice. The model is based on a two-stage network. Motion cues are obtained with a lightweight graph convolutional network that processes face landmarks. Then, both audio and motion features are fed to an audio-visual transformer which produces a fairly good estimation of the isolated target source. In a second stage, the predominant voice is enhanced with an audio-only network. We present different ablation studies and comparison to state-of-the-art methods. Finally, we explore the transferability of models trained for speech separation in the task of singing voice separation. The demos, code, and weights are available in https://ipcv.github.io/VoViT/"