Jérôme Revaud

Philippe Weinzaepfel, Thomas Lucas, Vincent Leroy et al.

Despite impressive performance for high-level downstream tasks, self-supervised pre-training methods have not yet fully delivered on dense geometric vision tasks such as stereo matching or optical flow. The application of self-supervised concepts, such as instance discrimination or masked image modeling, to geometric tasks is an active area of research. In this work, we build on the recent cross-view completion framework, a variation of masked image modeling that leverages a second view from the same scene which makes it well suited for binocular downstream tasks. The applicability of this concept has so far been limited in at least two ways: (a) by the difficulty of collecting real-world image pairs -- in practice only synthetic data have been used -- and (b) by the lack of generalization of vanilla transformers to dense downstream tasks for which relative position is more meaningful than absolute position. We explore three avenues of improvement. First, we introduce a method to collect suitable real-world image pairs at large scale. Second, we experiment with relative positional embeddings and show that they enable vision transformers to perform substantially better. Third, we scale up vision transformer based cross-completion architectures, which is made possible by the use of large amounts of data. With these improvements, we show for the first time that state-of-the-art results on stereo matching and optical flow can be reached without using any classical task-specific techniques like correlation volume, iterative estimation, image warping or multi-scale reasoning, thus paving the way towards universal vision models.

Philippe Weinzaepfel, Vincent Leroy, Thomas Lucas et al.

Masked Image Modeling (MIM) has recently been established as a potent pre-training paradigm. A pretext task is constructed by masking patches in an input image, and this masked content is then predicted by a neural network using visible patches as sole input. This pre-training leads to state-of-the-art performance when finetuned for high-level semantic tasks, e.g. image classification and object detection. In this paper we instead seek to learn representations that transfer well to a wide variety of 3D vision and lower-level geometric downstream tasks, such as depth prediction or optical flow estimation. Inspired by MIM, we propose an unsupervised representation learning task trained from pairs of images showing the same scene from different viewpoints. More precisely, we propose the pretext task of cross-view completion where the first input image is partially masked, and this masked content has to be reconstructed from the visible content and the second image. In single-view MIM, the masked content often cannot be inferred precisely from the visible portion only, so the model learns to act as a prior influenced by high-level semantics. In contrast, this ambiguity can be resolved with cross-view completion from the second unmasked image, on the condition that the model is able to understand the spatial relationship between the two images. Our experiments show that our pretext task leads to significantly improved performance for monocular 3D vision downstream tasks such as depth estimation. In addition, our model can be directly applied to binocular downstream tasks like optical flow or relative camera pose estimation, for which we obtain competitive results without bells and whistles, i.e., using a generic architecture without any task-specific design.

Jerome Revaud, Yohann Cabon, Romain Brégier et al.

Scene coordinates regression (SCR), i.e., predicting 3D coordinates for every pixel of a given image, has recently shown promising potential. However, existing methods remain limited to small scenes memorized during training, and thus hardly scale to realistic datasets and scenarios. In this paper, we propose a generalized SCR model trained once to be deployed in new test scenes, regardless of their scale, without any finetuning. Instead of encoding the scene coordinates into the network weights, our model takes as input a database image with some sparse 2D pixel to 3D coordinate annotations, extracted from e.g. off-the-shelf Structure-from-Motion or RGB-D data, and a query image for which are predicted a dense 3D coordinate map and its confidence, based on cross-attention. At test time, we rely on existing off-the-shelf image retrieval systems and fuse the predictions from a shortlist of relevant database images w.r.t. the query. Afterwards camera pose is obtained using standard Perspective-n-Point (PnP). Starting from selfsupervised CroCo pretrained weights, we train our model on diverse datasets to ensure generalizabilty across various scenarios, and significantly outperform other scene regression approaches, including scene-specific models, on multiple visual localization benchmarks. Finally, we show that the database representation of images and their 2D-3D annotations can be highly compressed with negligible loss of localization performance.

JongMin Lee, Yohann Cabon, Romain Brégier et al.

Existing learning-based methods for object pose estimation in RGB images are mostly model-specific or category based. They lack the capability to generalize to new object categories at test time, hence severely hindering their practicability and scalability. Notably, recent attempts have been made to solve this issue, but they still require accurate 3D data of the object surface at both train and test time. In this paper, we introduce a novel approach that can estimate in a single forward pass the pose of objects never seen during training, given minimum input. In contrast to existing state-of-the-art approaches, which rely on task-specific modules, our proposed model is entirely based on a transformer architecture, which can benefit from recently proposed 3D-geometry general pretraining. We conduct extensive experiments and report state-of-the-art one-shot performance on the challenging LINEMOD benchmark. Finally, extensive ablations allow us to determine good practices with this relatively new type of architecture in the field.

Vincent Leroy, Jerome Revaud, Thomas Lucas et al.

Transformers have become the standard in state-of-the-art vision architectures, achieving impressive performance on both image-level and dense pixelwise tasks. However, training vision transformers for high-resolution pixelwise tasks has a prohibitive cost. Typical solutions boil down to hierarchical architectures, fast and approximate attention, or training on low-resolution crops. This latter solution does not constrain architectural choices, but it leads to a clear performance drop when testing at resolutions significantly higher than that used for training, thus requiring ad-hoc and slow post-processing schemes. In this paper, we propose a novel strategy for efficient training and inference of high-resolution vision transformers. The key principle is to mask out most of the high-resolution inputs during training, keeping only N random windows. This allows the model to learn local interactions between tokens inside each window, and global interactions between tokens from different windows. As a result, the model can directly process the high-resolution input at test time without any special trick. We show that this strategy is effective when using relative positional embedding such as rotary embeddings. It is 4 times faster to train than a full-resolution network, and it is straightforward to use at test time compared to existing approaches. We apply this strategy to three dense prediction tasks with high-resolution data. First, we show on the task of semantic segmentation that a simple setting with 2 windows performs best, hence the name of our method: Win-Win. Second, we confirm this result on the task of monocular depth prediction. Third, we further extend it to the binocular task of optical flow, reaching state-of-the-art performance on the Spring benchmark that contains Full-HD images with an order of magnitude faster inference than the best competitor.

Leonid Antsfeld, Boris Chidlovskii, Yohann Cabon et al.

The recent paradigm shift in 3D vision led to the rise of foundation models with remarkable capabilities in 3D perception from uncalibrated images. However, extending these models to large-scale RGB stream 3D reconstruction remains challenging due to memory limitations. This work proposes S-MUSt3R, a simple and efficient pipeline that extends the limits of foundation models for monocular 3D reconstruction. Our approach addresses the scalability bottleneck of foundation models through a simple strategy of sequence segmentation followed by segment alignment and lightweight loop closure optimization. Without model retraining, we benefit from remarkable 3D reconstruction capacities of MUSt3R model and achieve trajectory and reconstruction performance comparable to traditional methods with more complex architecture. We evaluate S-MUSt3R on TUM, 7-Scenes and proprietary robot navigation datasets and show that S-MUSt3R runs successfully on long RGB sequences and produces accurate and consistent 3D reconstruction. Our results highlight the potential of leveraging the MUSt3R model for scalable monocular 3D scene in real-world settings, with an important advantage of making predictions directly in the metric space.

Shuzhe Wang, Vincent Leroy, Yohann Cabon et al.

Multi-view stereo reconstruction (MVS) in the wild requires to first estimate the camera parameters e.g. intrinsic and extrinsic parameters. These are usually tedious and cumbersome to obtain, yet they are mandatory to triangulate corresponding pixels in 3D space, which is the core of all best performing MVS algorithms. In this work, we take an opposite stance and introduce DUSt3R, a radically novel paradigm for Dense and Unconstrained Stereo 3D Reconstruction of arbitrary image collections, i.e. operating without prior information about camera calibration nor viewpoint poses. We cast the pairwise reconstruction problem as a regression of pointmaps, relaxing the hard constraints of usual projective camera models. We show that this formulation smoothly unifies the monocular and binocular reconstruction cases. In the case where more than two images are provided, we further propose a simple yet effective global alignment strategy that expresses all pairwise pointmaps in a common reference frame. We base our network architecture on standard Transformer encoders and decoders, allowing us to leverage powerful pretrained models. Our formulation directly provides a 3D model of the scene as well as depth information, but interestingly, we can seamlessly recover from it, pixel matches, relative and absolute camera. Exhaustive experiments on all these tasks showcase that the proposed DUSt3R can unify various 3D vision tasks and set new SoTAs on monocular/multi-view depth estimation as well as relative pose estimation. In summary, DUSt3R makes many geometric 3D vision tasks easy.

Martin Humenberger, Yohann Cabon, Nicolas Guerin et al.

Visual localization tackles the challenge of estimating the camera pose from images by using correspondence analysis between query images and a map. This task is computation and data intensive which poses challenges on thorough evaluation of methods on various datasets. However, in order to further advance in the field, we claim that robust visual localization algorithms should be evaluated on multiple datasets covering a broad domain variety. To facilitate this, we introduce kapture, a new, flexible, unified data format and toolbox for visual localization and structure-from-motion (SFM). It enables easy usage of different datasets as well as efficient and reusable data processing. To demonstrate this, we present a versatile pipeline for visual localization that facilitates the use of different local and global features, 3D data (e.g. depth maps), non-vision sensor data (e.g. IMU, GPS, WiFi), and various processing algorithms. Using multiple configurations of the pipeline, we show the great versatility of kapture in our experiments. Furthermore, we evaluate our methods on eight public datasets where they rank top on all and first on many of them. To foster future research, we release code, models, and all datasets used in this paper in the kapture format open source under a permissive BSD license. github.com/naver/kapture, github.com/naver/kapture-localization

Yohann Cabon, Lucas Stoffl, Leonid Antsfeld et al.

DUSt3R introduced a novel paradigm in geometric computer vision by proposing a model that can provide dense and unconstrained Stereo 3D Reconstruction of arbitrary image collections with no prior information about camera calibration nor viewpoint poses. Under the hood, however, DUSt3R processes image pairs, regressing local 3D reconstructions that need to be aligned in a global coordinate system. The number of pairs, growing quadratically, is an inherent limitation that becomes especially concerning for robust and fast optimization in the case of large image collections. In this paper, we propose an extension of DUSt3R from pairs to multiple views, that addresses all aforementioned concerns. Indeed, we propose a Multi-view Network for Stereo 3D Reconstruction, or MUSt3R, that modifies the DUSt3R architecture by making it symmetric and extending it to directly predict 3D structure for all views in a common coordinate frame. Second, we entail the model with a multi-layer memory mechanism which allows to reduce the computational complexity and to scale the reconstruction to large collections, inferring thousands of 3D pointmaps at high frame-rates with limited added complexity. The framework is designed to perform 3D reconstruction both offline and online, and hence can be seamlessly applied to SfM and visual SLAM scenarios showing state-of-the-art performance on various 3D downstream tasks, including uncalibrated Visual Odometry, relative camera pose, scale and focal estimation, 3D reconstruction and multi-view depth estimation.

Wonbong Jang, Philippe Weinzaepfel, Vincent Leroy et al.

We present Pow3r, a novel large 3D vision regression model that is highly versatile in the input modalities it accepts. Unlike previous feed-forward models that lack any mechanism to exploit known camera or scene priors at test time, Pow3r incorporates any combination of auxiliary information such as intrinsics, relative pose, dense or sparse depth, alongside input images, within a single network. Building upon the recent DUSt3R paradigm, a transformer-based architecture that leverages powerful pre-training, our lightweight and versatile conditioning acts as additional guidance for the network to predict more accurate estimates when auxiliary information is available. During training we feed the model with random subsets of modalities at each iteration, which enables the model to operate under different levels of known priors at test time. This in turn opens up new capabilities, such as performing inference in native image resolution, or point-cloud completion. Our experiments on 3D reconstruction, depth completion, multi-view depth prediction, multi-view stereo, and multi-view pose estimation tasks yield state-of-the-art results and confirm the effectiveness of Pow3r at exploiting all available information. The project webpage is https://europe.naverlabs.com/pow3r.

Vincent Leroy, Yohann Cabon, Jérôme Revaud

Image Matching is a core component of all best-performing algorithms and pipelines in 3D vision. Yet despite matching being fundamentally a 3D problem, intrinsically linked to camera pose and scene geometry, it is typically treated as a 2D problem. This makes sense as the goal of matching is to establish correspondences between 2D pixel fields, but also seems like a potentially hazardous choice. In this work, we take a different stance and propose to cast matching as a 3D task with DUSt3R, a recent and powerful 3D reconstruction framework based on Transformers. Based on pointmaps regression, this method displayed impressive robustness in matching views with extreme viewpoint changes, yet with limited accuracy. We aim here to improve the matching capabilities of such an approach while preserving its robustness. We thus propose to augment the DUSt3R network with a new head that outputs dense local features, trained with an additional matching loss. We further address the issue of quadratic complexity of dense matching, which becomes prohibitively slow for downstream applications if not carefully treated. We introduce a fast reciprocal matching scheme that not only accelerates matching by orders of magnitude, but also comes with theoretical guarantees and, lastly, yields improved results. Extensive experiments show that our approach, coined MASt3R, significantly outperforms the state of the art on multiple matching tasks. In particular, it beats the best published methods by 30% (absolute improvement) in VCRE AUC on the extremely challenging Map-free localization dataset.

Shyamgopal Karthik, Jérome Revaud, Boris Chidlovskii

Class imbalance and noisy labels are the norm rather than the exception in many large-scale classification datasets. Nevertheless, most works in machine learning typically assume balanced and clean data. There have been some recent attempts to tackle, on one side, the problem of learning from noisy labels and, on the other side, learning from long-tailed data. Each group of methods make simplifying assumptions about the other. Due to this separation, the proposed solutions often underperform when both assumptions are violated. In this work, we present a simple two-stage approach based on recent advances in self-supervised learning to treat both challenges simultaneously. It consists of, first, task-agnostic self-supervised pre-training, followed by task-specific fine-tuning using an appropriate loss. Most significantly, we find that self-supervised learning approaches are effectively able to cope with severe class imbalance. In addition, the resulting learned representations are also remarkably robust to label noise, when fine-tuned with an imbalance- and noise-resistant loss function. We validate our claims with experiments on CIFAR-10 and CIFAR-100 augmented with synthetic imbalance and noise, as well as the large-scale inherently noisy Clothing-1M dataset.

Jerome Revaud, Jon Almazan, Rafael Sampaio de Rezende et al.

Image retrieval can be formulated as a ranking problem where the goal is to order database images by decreasing similarity to the query. Recent deep models for image retrieval have outperformed traditional methods by leveraging ranking-tailored loss functions, but important theoretical and practical problems remain. First, rather than directly optimizing the global ranking, they minimize an upper-bound on the essential loss, which does not necessarily result in an optimal mean average precision (mAP). Second, these methods require significant engineering efforts to work well, e.g. special pre-training and hard-negative mining. In this paper we propose instead to directly optimize the global mAP by leveraging recent advances in listwise loss formulations. Using a histogram binning approximation, the AP can be differentiated and thus employed to end-to-end learning. Compared to existing losses, the proposed method considers thousands of images simultaneously at each iteration and eliminates the need for ad hoc tricks. It also establishes a new state of the art on many standard retrieval benchmarks. Models and evaluation scripts have been made available at https://europe.naverlabs.com/Deep-Image-Retrieval/

Jerome Revaud, Philippe Weinzaepfel, César De Souza et al.

Interest point detection and local feature description are fundamental steps in many computer vision applications. Classical methods for these tasks are based on a detect-then-describe paradigm where separate handcrafted methods are used to first identify repeatable keypoints and then represent them with a local descriptor. Neural networks trained with metric learning losses have recently caught up with these techniques, focusing on learning repeatable saliency maps for keypoint detection and learning descriptors at the detected keypoint locations. In this work, we argue that salient regions are not necessarily discriminative, and therefore can harm the performance of the description. Furthermore, we claim that descriptors should be learned only in regions for which matching can be performed with high confidence. We thus propose to jointly learn keypoint detection and description together with a predictor of the local descriptor discriminativeness. This allows us to avoid ambiguous areas and leads to reliable keypoint detections and descriptions. Our detection-and-description approach, trained with self-supervision, can simultaneously output sparse, repeatable and reliable keypoints that outperforms state-of-the-art detectors and descriptors on the HPatches dataset. It also establishes a record on the recently released Aachen Day-Night localization dataset.

Albert Gordo, Jon Almazan, Jerome Revaud et al.

While deep learning has become a key ingredient in the top performing methods for many computer vision tasks, it has failed so far to bring similar improvements to instance-level image retrieval. In this article, we argue that reasons for the underwhelming results of deep methods on image retrieval are threefold: i) noisy training data, ii) inappropriate deep architecture, and iii) suboptimal training procedure. We address all three issues. First, we leverage a large-scale but noisy landmark dataset and develop an automatic cleaning method that produces a suitable training set for deep retrieval. Second, we build on the recent R-MAC descriptor, show that it can be interpreted as a deep and differentiable architecture, and present improvements to enhance it. Last, we train this network with a siamese architecture that combines three streams with a triplet loss. At the end of the training process, the proposed architecture produces a global image representation in a single forward pass that is well suited for image retrieval. Extensive experiments show that our approach significantly outperforms previous retrieval approaches, including state-of-the-art methods based on costly local descriptor indexing and spatial verification. On Oxford 5k, Paris 6k and Holidays, we respectively report 94.7, 96.6, and 94.8 mean average precision. Our representations can also be heavily compressed using product quantization with little loss in accuracy. For additional material, please see www.xrce.xerox.com/Deep-Image-Retrieval.

Albert Gordo, Jon Almazan, Jerome Revaud et al.

We propose a novel approach for instance-level image retrieval. It produces a global and compact fixed-length representation for each image by aggregating many region-wise descriptors. In contrast to previous works employing pre-trained deep networks as a black box to produce features, our method leverages a deep architecture trained for the specific task of image retrieval. Our contribution is twofold: (i) we leverage a ranking framework to learn convolution and projection weights that are used to build the region features; and (ii) we employ a region proposal network to learn which regions should be pooled to form the final global descriptor. We show that using clean training data is key to the success of our approach. To that aim, we use a large scale but noisy landmark dataset and develop an automatic cleaning approach. The proposed architecture produces a global image representation in a single forward pass. Our approach significantly outperforms previous approaches based on global descriptors on standard datasets. It even surpasses most prior works based on costly local descriptor indexing and spatial verification. Additional material is available at www.xrce.xerox.com/Deep-Image-Retrieval.

Danila Potapov, Matthijs Douze, Jerome Revaud et al.

While important advances were recently made towards temporally localizing and recognizing specific human actions or activities in videos, efficient detection and classification of long video chunks belonging to semantically defined categories such as "pursuit" or "romance" remains challenging.We introduce a new dataset, Action Movie Franchises, consisting of a collection of Hollywood action movie franchises. We define 11 non-exclusive semantic categories - called beat-categories - that are broad enough to cover most of the movie footage. The corresponding beat-events are annotated as groups of video shots, possibly overlapping.We propose an approach for localizing beat-events based on classifying shots into beat-categories and learning the temporal constraints between shots. We show that temporal constraints significantly improve the classification performance. We set up an evaluation protocol for beat-event localization as well as for shot classification, depending on whether movies from the same franchise are present or not in the training data.

Jerome Revaud, Philippe Weinzaepfel, Zaid Harchaoui et al.

We introduce a novel matching algorithm, called DeepMatching, to compute dense correspondences between images. DeepMatching relies on a hierarchical, multi-layer, correlational architecture designed for matching images and was inspired by deep convolutional approaches. The proposed matching algorithm can handle non-rigid deformations and repetitive textures and efficiently determines dense correspondences in the presence of significant changes between images. We evaluate the performance of DeepMatching, in comparison with state-of-the-art matching algorithms, on the Mikolajczyk (Mikolajczyk et al 2005), the MPI-Sintel (Butler et al 2012) and the Kitti (Geiger et al 2013) datasets. DeepMatching outperforms the state-of-the-art algorithms and shows excellent results in particular for repetitive textures.We also propose a method for estimating optical flow, called DeepFlow, by integrating DeepMatching in the large displacement optical flow (LDOF) approach of Brox and Malik (2011). Compared to existing matching algorithms, additional robustness to large displacements and complex motion is obtained thanks to our matching approach. DeepFlow obtains competitive performance on public benchmarks for optical flow estimation.

Matthijs Douze, Jérôme Revaud, Jakob Verbeek et al.

We address the problem of specific video event retrieval. Given a query video of a specific event, e.g., a concert of Madonna, the goal is to retrieve other videos of the same event that temporally overlap with the query. Our approach encodes the frame descriptors of a video to jointly represent their appearance and temporal order. It exploits the properties of circulant matrices to efficiently compare the videos in the frequency domain. This offers a significant gain in complexity and accurately localizes the matching parts of videos. The descriptors can be compressed in the frequency domain with a product quantizer adapted to complex numbers. In this case, video retrieval is performed without decompressing the descriptors. We also consider the temporal alignment of a set of videos. We exploit the matching confidence and an estimate of the temporal offset computed for all pairs of videos by our retrieval approach. Our robust algorithm aligns the videos on a global timeline by maximizing the set of temporally consistent matches. The global temporal alignment enables synchronous playback of the videos of a given scene.

Jerome Revaud, Philippe Weinzaepfel, Zaid Harchaoui et al.

We propose a novel approach for optical flow estimation , targeted at large displacements with significant oc-clusions. It consists of two steps: i) dense matching by edge-preserving interpolation from a sparse set of matches; ii) variational energy minimization initialized with the dense matches. The sparse-to-dense interpolation relies on an appropriate choice of the distance, namely an edge-aware geodesic distance. This distance is tailored to handle occlusions and motion boundaries -- two common and difficult issues for optical flow computation. We also propose an approximation scheme for the geodesic distance to allow fast computation without loss of performance. Subsequent to the dense interpolation step, standard one-level variational energy minimization is carried out on the dense matches to obtain the final flow estimation. The proposed approach, called Edge-Preserving Interpolation of Correspondences (EpicFlow) is fast and robust to large displacements. It significantly outperforms the state of the art on MPI-Sintel and performs on par on Kitti and Middlebury.