Mihai Dusmanu

Hao Dong, Xieyuanli Chen, Mihai Dusmanu et al.

A distinctive representation of image patches in form of features is a key component of many computer vision and robotics tasks, such as image matching, image retrieval, and visual localization. State-of-the-art descriptors, from hand-crafted descriptors such as SIFT to learned ones such as HardNet, are usually high dimensional; 128 dimensions or even more. The higher the dimensionality, the larger the memory consumption and computational time for approaches using such descriptors. In this paper, we investigate multi-layer perceptrons (MLPs) to extract low-dimensional but high-quality descriptors. We thoroughly analyze our method in unsupervised, self-supervised, and supervised settings, and evaluate the dimensionality reduction results on four representative descriptors. We consider different applications, including visual localization, patch verification, image matching and retrieval. The experiments show that our lightweight MLPs achieve better dimensionality reduction than PCA. The lower-dimensional descriptors generated by our approach outperform the original higher-dimensional descriptors in downstream tasks, especially for the hand-crafted ones. The code will be available at https://github.com/PRBonn/descriptor-dr.

Paul-Edouard Sarlin, Mihai Dusmanu, Johannes L. Schönberger et al.

Localization and mapping is the foundational technology for augmented reality (AR) that enables sharing and persistence of digital content in the real world. While significant progress has been made, researchers are still mostly driven by unrealistic benchmarks not representative of real-world AR scenarios. These benchmarks are often based on small-scale datasets with low scene diversity, captured from stationary cameras, and lack other sensor inputs like inertial, radio, or depth data. Furthermore, their ground-truth (GT) accuracy is mostly insufficient to satisfy AR requirements. To close this gap, we introduce LaMAR, a new benchmark with a comprehensive capture and GT pipeline that co-registers realistic trajectories and sensor streams captured by heterogeneous AR devices in large, unconstrained scenes. To establish an accurate GT, our pipeline robustly aligns the trajectories against laser scans in a fully automated manner. As a result, we publish a benchmark dataset of diverse and large-scale scenes recorded with head-mounted and hand-held AR devices. We extend several state-of-the-art methods to take advantage of the AR-specific setup and evaluate them on our benchmark. The results offer new insights on current research and reveal promising avenues for future work in the field of localization and mapping for AR.

Sayan Deb Sarkar, Rémi Pautrat, Ondrej Miksik et al. · stanford

Video Language Models (VideoLMs) empower AI systems to understand temporal dynamics in videos. To fit to the maximum context window constraint, current methods use keyframe sampling which can miss both macro-level events and micro-level details due to the sparse temporal coverage. Furthermore, processing full images and their tokens for each frame incurs substantial computational overhead. To address these limitations, we propose to leverage video codec primitives (specifically motion vectors and residuals) which natively encode video redundancy and sparsity without requiring expensive full-image encoding for most frames. To this end, we introduce lightweight transformer-based encoders that aggregate codec primitives and align their representations with image encoder embeddings through a pre-training strategy that accelerates convergence during end-to-end fine-tuning. Our approach reduces the time-to-first-token by up to $86\%$ and token usage by up to $93\%$ compared to standard VideoLMs. Moreover, by varying the keyframe and codec primitive densities we are able to maintain or exceed performance on $14$ diverse video understanding benchmarks spanning general question answering, temporal reasoning, long-form understanding, and spatial scene understanding.

Emilia Szymanska, Mihai Dusmanu, Jan-Willem Buurlage et al.

Answering questions about the spatial properties of the environment poses challenges for existing language and vision foundation models due to a lack of understanding of the 3D world notably in terms of relationships between objects. To push the field forward, multiple 3D Q&A datasets were proposed which, overall, provide a variety of questions, but they individually focus on particular aspects of 3D reasoning or are limited in terms of data modalities. To address this, we present Space3D-Bench - a collection of 1000 general spatial questions and answers related to scenes of the Replica dataset which offers a variety of data modalities: point clouds, posed RGB-D images, navigation meshes and 3D object detections. To ensure that the questions cover a wide range of 3D objectives, we propose an indoor spatial questions taxonomy inspired by geographic information systems and use it to balance the dataset accordingly. Moreover, we provide an assessment system that grades natural language responses based on predefined ground-truth answers by leveraging a Vision Language Model's comprehension of both text and images to compare the responses with ground-truth textual information or relevant visual data. Finally, we introduce a baseline called RAG3D-Chat integrating the world understanding of foundation models with rich context retrieval, achieving an accuracy of 67% on the proposed dataset.

Lukas Bösiger, Mihai Dusmanu, Marc Pollefeys et al.

Rendering realistic images from 3D reconstruction is an essential task of many Computer Vision and Robotics pipelines, notably for mixed-reality applications as well as training autonomous agents in simulated environments. However, the quality of novel views heavily depends of the source reconstruction which is often imperfect due to noisy or missing geometry and appearance. Inspired by the recent success of reference-based super-resolution networks, we propose MaRINeR, a refinement method that leverages information of a nearby mapping image to improve the rendering of a target viewpoint. We first establish matches between the raw rendered image of the scene geometry from the target viewpoint and the nearby reference based on deep features, followed by hierarchical detail transfer. We show improved renderings in quantitative metrics and qualitative examples from both explicit and implicit scene representations. We further employ our method on the downstream tasks of pseudo-ground-truth validation, synthetic data enhancement and detail recovery for renderings of reduced 3D reconstructions.

Kevin Qu, Haozhe Qi, Mihai Dusmanu et al.

Multimodal Large Language Models (MLLMs) have made impressive progress in connecting vision and language, but they still struggle with spatial understanding and viewpoint-aware reasoning. Recent efforts aim to augment the input representations with geometric cues rather than explicitly teaching models to reason in 3D space. We introduce Loc3R-VLM, a framework that equips 2D Vision-Language Models with advanced 3D understanding capabilities from monocular video input. Inspired by human spatial cognition, Loc3R-VLM relies on two joint objectives: global layout reconstruction to build a holistic representation of the scene structure, and explicit situation modeling to anchor egocentric perspective. These objectives provide direct spatial supervision that grounds both perception and language in a 3D context. To ensure geometric consistency and metric-scale alignment, we leverage lightweight camera pose priors extracted from a pre-trained 3D foundation model. Loc3R-VLM achieves state-of-the-art performance in language-based localization and outperforms existing 2D- and video-based approaches on situated and general 3D question-answering benchmarks, demonstrating that our spatial supervision framework enables strong 3D understanding. Project page: https://kevinqu7.github.io/loc3r-vlm

Gen Li, Kaifeng Zhao, Siwei Zhang et al.

Understanding the world in first-person view is fundamental in Augmented Reality (AR). This immersive perspective brings dramatic visual changes and unique challenges compared to third-person views. Synthetic data has empowered third-person-view vision models, but its application to embodied egocentric perception tasks remains largely unexplored. A critical challenge lies in simulating natural human movements and behaviors that effectively steer the embodied cameras to capture a faithful egocentric representation of the 3D world. To address this challenge, we introduce EgoGen, a new synthetic data generator that can produce accurate and rich ground-truth training data for egocentric perception tasks. At the heart of EgoGen is a novel human motion synthesis model that directly leverages egocentric visual inputs of a virtual human to sense the 3D environment. Combined with collision-avoiding motion primitives and a two-stage reinforcement learning approach, our motion synthesis model offers a closed-loop solution where the embodied perception and movement of the virtual human are seamlessly coupled. Compared to previous works, our model eliminates the need for a pre-defined global path, and is directly applicable to dynamic environments. Combined with our easy-to-use and scalable data generation pipeline, we demonstrate EgoGen's efficacy in three tasks: mapping and localization for head-mounted cameras, egocentric camera tracking, and human mesh recovery from egocentric views. EgoGen will be fully open-sourced, offering a practical solution for creating realistic egocentric training data and aiming to serve as a useful tool for egocentric computer vision research. Refer to our project page: https://ego-gen.github.io/.

Arda Düzçeker, Silvano Galliani, Christoph Vogel et al.

We propose an online multi-view depth prediction approach on posed video streams, where the scene geometry information computed in the previous time steps is propagated to the current time step in an efficient and geometrically plausible way. The backbone of our approach is a real-time capable, lightweight encoder-decoder that relies on cost volumes computed from pairs of images. We extend it by placing a ConvLSTM cell at the bottleneck layer, which compresses an arbitrary amount of past information in its states. The novelty lies in propagating the hidden state of the cell by accounting for the viewpoint changes between time steps. At a given time step, we warp the previous hidden state into the current camera plane using the previous depth prediction. Our extension brings only a small overhead of computation time and memory consumption, while improving the depth predictions significantly. As a result, we outperform the existing state-of-the-art multi-view stereo methods on most of the evaluated metrics in hundreds of indoor scenes while maintaining a real-time performance. Code available: https://github.com/ardaduz/deep-video-mvs

Nicolas Pfitzer, Yifan Zhou, Marco Poggensee et al.

Over 43 million people worldwide live with severe visual impairment, facing significant challenges in navigating unfamiliar environments. We present MR.NAVI, a mixed reality system that enhances spatial awareness for visually impaired users through real-time scene understanding and intuitive audio feedback. Our system combines computer vision algorithms for object detection and depth estimation with natural language processing to provide contextual scene descriptions, proactive collision avoidance, and navigation instructions. The distributed architecture processes sensor data through MobileNet for object detection and employs RANSAC-based floor detection with DBSCAN clustering for obstacle avoidance. Integration with public transit APIs enables navigation with public transportation directions. Through our experiments with user studies, we evaluated both scene description and navigation functionalities in unfamiliar environments, showing promising usability and effectiveness.

Mihai Dusmanu, Ondrej Miksik, Johannes L. Schönberger et al.

Visual localization and mapping is the key technology underlying the majority of mixed reality and robotics systems. Most state-of-the-art approaches rely on local features to establish correspondences between images. In this paper, we present three novel scenarios for localization and mapping which require the continuous update of feature representations and the ability to match across different feature types. While localization and mapping is a fundamental computer vision problem, the traditional setup supposes the same local features are used throughout the evolution of a map. Thus, whenever the underlying features are changed, the whole process is repeated from scratch. However, this is typically impossible in practice, because raw images are often not stored and re-building the maps could lead to loss of the attached digital content. To overcome the limitations of current approaches, we present the first principled solution to cross-descriptor localization and mapping. Our data-driven approach is agnostic to the feature descriptor type, has low computational requirements, and scales linearly with the number of description algorithms. Extensive experiments demonstrate the effectiveness of our approach on state-of-the-art benchmarks for a variety of handcrafted and learned features.

Mihai Dusmanu, Johannes L. Schönberger, Sudipta N. Sinha et al.

Many computer vision systems require users to upload image features to the cloud for processing and storage. These features can be exploited to recover sensitive information about the scene or subjects, e.g., by reconstructing the appearance of the original image. To address this privacy concern, we propose a new privacy-preserving feature representation. The core idea of our work is to drop constraints from each feature descriptor by embedding it within an affine subspace containing the original feature as well as adversarial feature samples. Feature matching on the privacy-preserving representation is enabled based on the notion of subspace-to-subspace distance. We experimentally demonstrate the effectiveness of our method and its high practical relevance for the applications of visual localization and mapping as well as face authentication. Compared to the original features, our approach makes it significantly more difficult for an adversary to recover private information.

Mihai Dusmanu, Johannes L. Schönberger, Marc Pollefeys

In this work, we address the problem of refining the geometry of local image features from multiple views without known scene or camera geometry. Current approaches to local feature detection are inherently limited in their keypoint localization accuracy because they only operate on a single view. This limitation has a negative impact on downstream tasks such as Structure-from-Motion, where inaccurate keypoints lead to large errors in triangulation and camera localization. Our proposed method naturally complements the traditional feature extraction and matching paradigm. We first estimate local geometric transformations between tentative matches and then optimize the keypoint locations over multiple views jointly according to a non-linear least squares formulation. Throughout a variety of experiments, we show that our method consistently improves the triangulation and camera localization performance for both hand-crafted and learned local features.

Mihai Dusmanu, Ignacio Rocco, Tomas Pajdla et al.

In this work we address the problem of finding reliable pixel-level correspondences under difficult imaging conditions. We propose an approach where a single convolutional neural network plays a dual role: It is simultaneously a dense feature descriptor and a feature detector. By postponing the detection to a later stage, the obtained keypoints are more stable than their traditional counterparts based on early detection of low-level structures. We show that this model can be trained using pixel correspondences extracted from readily available large-scale SfM reconstructions, without any further annotations. The proposed method obtains state-of-the-art performance on both the difficult Aachen Day-Night localization dataset and the InLoc indoor localization benchmark, as well as competitive performance on other benchmarks for image matching and 3D reconstruction.