| CARVIEW |
Select Language
HTTP/2 200
server: GitHub.com
content-type: text/html; charset=utf-8
last-modified: Tue, 23 Dec 2025 05:04:51 GMT
access-control-allow-origin: *
etag: W/"694a22f3-ca1d"
expires: Sun, 28 Dec 2025 02:35:40 GMT
cache-control: max-age=600
content-encoding: gzip
x-proxy-cache: MISS
x-github-request-id: 9B9A:3655F2:72F326:809978:69509524
accept-ranges: bytes
age: 0
date: Sun, 28 Dec 2025 02:25:40 GMT
via: 1.1 varnish
x-served-by: cache-bom-vanm7210083-BOM
x-cache: MISS
x-cache-hits: 0
x-timer: S1766888740.384587,VS0,VE204
vary: Accept-Encoding
x-fastly-request-id: 21d55c25bf82e2b727f0083b30853ce635b9e30e
content-length: 12154
Eldar Insafutdinov
Eldar Insafutdinov
I am a computer vision researcher working on 3D perception at the Amazon Frontier AI and Robotics (FAR). Formerly, I was a postdoc in the Visual Geometry Group at the University of Oxford working with Prof. Andrea Vedaldi and Dr. Joao Henriques on self-supervised learning of visual representations for novel view synthesis and acquisition of 3D assets from in-the-wild data. I completed my PhD at the Max Planck Institute for Informatics under the supervision of Prof. Bernt Schiele working on human pose estimation. I had the pleasure to do an internship at the Intelligent Systems Lab at Intel working with Alexey Dosovitskiy. I obtained my master’s degree in Visual Computing from the Saarland University.
Contact | Google Scholar | GitHub | LinkedIn
Publications
Dynamic Point Maps: A Versatile Representation for Dynamic 3D Reconstruction
Edgar Sucar,
Zihang Lai,
Eldar Insafutdinov,
Andrea Vedaldi
ICCV, 2025
DUSt3R has recently shown that one can reduce many tasks in multi-view geometry, including estimating camera intrinsics and extrinsics, reconstructing the scene in 3D, and establishing image correspondences, to the prediction of a pair of viewpoint-invariant point maps, i.e., pixel-aligned point clouds defined in a common reference frame. This formulation is elegant and powerful, but unable to tackle dynamic scenes. To address this challenge, we introduce the concept of Dynamic Point Maps (DPM), extending standard point maps to support 4D tasks such as motion segmentation, scene flow estimation, 3D object tracking, and 2D correspondence. Our key intuition is that, when time is introduced, there are several possible spatial and time references that can be used to define the point maps. We identify a minimal subset of such combinations that can be regressed by a network to solve the sub tasks mentioned above. We train a DPM predictor on a mixture of synthetic and real data and evaluate it across diverse benchmarks for video depth prediction, dynamic point cloud reconstruction, 3D scene flow and object pose tracking, achieving state-of-the-art performance.
SEED4D: A Synthetic Ego–Exo Dynamic 4D Data Generator, Driving Dataset and Benchmark
Marius Kästingschäfer,
Théo Gieruc,
Sebastian Bernhard,
Dylan Campbell,
Eldar Insafutdinov,
Eyvaz Najafli,
Thomas Brox
to appear in WACV, 2025
Models for egocentric 3D and 4D reconstruction, including few-shot interpolation and extrapolation settings, can benefit from having images from exocentric viewpoints as supervision signals. No existing dataset provides the necessary mixture of complex, dynamic, and multi-view data. To facilitate the development of 3D and 4D reconstruction methods in the autonomous driving context, we propose a Synthetic Ego–Exo Dynamic 4D (SEED4D) data generator and dataset. We present a customizable, easy-to-use data generator for spatio-temporal multi-view data creation. Our open-source data generator allows the creation of synthetic data for camera setups commonly used in the NuScenes, KITTI360, and Waymo datasets. Additionally, SEED4D encompasses two large-scale multi-view synthetic urban scene datasets. Our static (3D) dataset encompasses 212k inward- and outward-facing vehicle images from 2k scenes, while our dynamic (4D) dataset contains 16.8M images from 10k trajectories, each sampled at 100 points in time with egocentric images, exocentric images, and LiDAR data.
Flash3D: Feed-Forward Generalisable 3D Scene Reconstruction from a Single Image
Stanislaw Szymanowicz,
Eldar Insafutdinov,
Chuanxia Zheng,
Dylan Campbell,
João F. Henriques,
Christian Rupprecht,
Andrea Vedaldi
to appear in 3DV, 2025
In this paper, we propose Flash3D, a method for scene reconstruction and novel view synthesis from a single image which is both very generalisable and efficient. For generalisability, we start from a “foundation” model for monocular depth estimation and extend it to a full 3D shape and appearance reconstructor. For efficiency, we base this extension on feed-forward Gaussian Splatting. Specifically, we predict a first layer of 3D Gaussians at the predicted depth, and then add additional layers of Gaussians that are offset in space, allowing the model to complete the reconstruction behind occlusions and truncations. Flash3D is very efficient, trainable on a single GPU in a day, and thus accessible to most researchers. It achieves state-of-the-art results when trained and tested on RealEstate10k. When transferred to unseen datasets like NYU it outperforms competitors by a large margin. More impressively, when transferred to KITTI, Flash3D achieves better PSNR than methods trained specifically on that dataset. In some instances, it even outperforms recent methods that use multiple views as input.
Neural fields for co-reconstructing 3D objects from incidental 2D data
Dylan Campbell,
Eldar Insafutdinov,
João F. Henriques,
Andrea Vedaldi
CVPR Workshops, 2024
We ask whether 3D objects can be reconstructed from real world data collected for some other purpose, such as autonomous driving or augmented reality, thus inferring objects only incidentally. 3D reconstruction from incidental data is a major challenge because, in addition to significant noise, only a few views of each object are observed, which are insufficient for reconstruction. We approach this problem as a co-reconstruction task, where multiple objects are reconstructed together, learning shape and appearance priors for regularization. In order to do so, we introduce a neural radiance field that is conditioned via an attention mechanism on the identity of the individual objects. We further disentangle shape from appearance and diffuse color from specular color via an asymmetric two-stream network, which factors shared information from instance-specific details. We demonstrate the ability of this method to reconstruct full 3D objects from partial, incidental observations in autonomous driving and other datasets.
SNeS: Learning Probably Symmetric Neural Surfaces from Incomplete Data
Eldar Insafutdinov,
Dylan Campbell,
João F. Henriques,
Andrea Vedaldi
ECCV, 2022
We present a method for the accurate 3D reconstruction of partly-symmetric objects. We build on the strengths of recent advances in neural reconstruction and rendering such as Neural Radiance Fields (NeRF). A major shortcoming of such approaches is that they fail to reconstruct any part of the object which is not clearly visible in the training image, which is often the case for in-the-wild images and videos. When evidence is lacking, structural priors such as symmetry can be used to complete the missing information. However, exploiting such priors in neural rendering is highly non-trivial: while geometry and non-reflective materials may be symmetric, shadows and reflections from the ambient scene are not symmetric in general. To address this, we apply a soft symmetry constraint to the 3D geometry and material properties, having factored appearance into lighting, albedo colour and reflectivity. We evaluate our method on the recently introduced CO3D dataset, focusing on the car category due to the challenge of reconstructing highly-reflective materials. We show that it can reconstruct unobserved regions with high fidelity and render high-quality novel view images.
Lifting 2D Object Locations to 3D by Discounting LiDAR Outliers across Objects and Views
Robert McCraith,
Eldar Insafutdinov,
Lukas Neumann,
Andrea Vedaldi
ICRA, 2022
We present a system for automatic converting of 2D mask object predictions and raw LiDAR point clouds into full 3D bounding boxes of objects. Because the LiDAR point clouds are partial, directly fitting bounding boxes to the point clouds is meaningless. Instead, we suggest that obtaining good results requires sharing information between all objects in the dataset jointly, over multiple frames. We then make three improvements to the baseline. First, we address ambiguities in predicting the object rotations via direct optimization in this space while still backpropagating rotation prediction through the model. Second, we explicitly model outliers and task the network with learning their typical patterns, thus better discounting them. Third, we enforce temporal consistency when video data is available. With these contributions, our method significantly outperforms previous work despite the fact that those methods use significantly more complex pipelines, 3D models and additional human-annotated external sources of prior information.
360-Degree Textures of People in Clothing from a Single Image
Verica Lazova,
Eldar Insafutdinov,
Verica Lazova
3DV, 2019
In this paper we predict a full 3D avatar of a person from a single image. We infer texture and geometry in the UV-space of the SMPL model using an image-to-image translation method. Given partial texture and segmentation layout maps derived from the input view, our model predicts the complete segmentation map, the complete texture map, and a displacement map. The predicted maps can be applied to the SMPL model in order to naturally generalize to novel poses, shapes, and even new clothing. In order to learn our model in a common UV-space, we non-rigidly register the SMPL model to thousands of 3D scans, effectively encoding textures and geometries as images in correspondence. This turns a difficult 3D inference task into a simpler image-to-image translation one. Results on rendered scans of people and images from the DeepFashion dataset demonstrate that our method can reconstruct plausible 3D avatars from a single image. We further use our model to digitally change pose, shape, swap garments between people and edit clothing. To encourage research in this direction we will make the source code available for research purpose.
Unsupervised Learning of Shape and Pose with Differentiable Point Clouds
Eldar Insafutdinov,
Alexey Dosovitskiy
NeurIPS, 2018
We address the problem of learning accurate 3D shape and camera pose from a collection of unlabeled category-specific images. We train a convolutional network to predict both the shape and the pose from a single image by minimizing the reprojection error: given several views of an object, the projections of the predicted shapes to the predicted camera poses should match the provided views. To deal with pose ambiguity, we introduce an ensemble of pose predictors that we then distill it to a single “student” model. To allow for efficient learning of high-fidelity shapes, we represent the shapes by point clouds and devise a formulation allowing for differentiable projection of these. Our experiments show that the distilled ensemble of pose predictors learns to estimate the pose accurately, while the point cloud representation allows to predict detailed shape models.
PoseTrack: A Benchmark for Human Pose Estimation and Tracking
Mykhaylo Andriluka,
Umar Iqbal,
Eldar Insafutdinov,
Leonid Pishchulin,
Anton Milan,
Juergen Gall,
Bernt Schiele
CVPR, 2018
Existing systems for video-based pose estimation and tracking struggle to perform well on realistic videos with multiple people and often fail to output body-pose trajectories consistent over time. To address this shortcoming this paper introduces PoseTrack which is a new large-scale benchmark for video-based human pose estimation and articulated tracking. Our new benchmark encompasses three tasks focusing on i) single-frame multi-person pose estimation, ii) multi-person pose estimation in videos, and iii) multi-person articulated tracking. To establish the benchmark, we collect, annotate and release a new dataset that features videos with multiple people labeled with person tracks and articulated pose. A public centralized evaluation server is provided to allow the research community to evaluate on a held-out test set. Furthermore, we conduct an extensive experimental study on recent approaches to articulated pose tracking and provide analysis of the strengths and weaknesses of the state of the art. We envision that the proposed benchmark will stimulate productive research both by providing a large and representative training dataset as well as providing a platform to objectively evaluate and compare the proposed methods.
ArtTrack: Articulated Multi-Person Tracking in the Wild
Eldar Insafutdinov,
Mykhaylo Andriluka,
Leonid Pishchulin,
Siyu Tang,
Evgeny Levinkov,
Bjoern Andres,
Bernt Schiele
CVPR, 2017 (oral presentation)
In this paper we propose an approach for articulated tracking of multiple people in unconstrained videos. Our starting point is a model that resembles existing architectures for single-frame pose estimation but is substantially faster. We achieve this in two ways: (1) by simplifying and sparsifying the body-part relationship graph and leveraging recent methods for faster inference, and (2) by offloading a substantial share of computation onto a feed-forward convolutional architecture that is able to detect and associate body joints of the same person even in clutter. We use this model to generate proposals for body joint locations and formulate articulated tracking as spatio-temporal grouping of such proposals. This allows to jointly solve the association problem for all people in the scene by propagating evidence from strong detections through time and enforcing constraints that each proposal can be assigned to one person only. We report results on a public “MPII Human Pose” benchmark and on a new “MPII Video Pose” dataset of image sequences with multiple people. We demonstrate that our model achieves state-of-the-art results while using only a fraction of time and is able to leverage temporal information to improve state-of-the-art for crowded scenes
Other
DeepLabCut
Above: courtesy of the Murthy (mouse), Leventhal (rat), and Axel (fly) labs.
My keypoint detection algorithm from the DeeperCut paper and its implementation served as the foundation for DeepLabCut, a toolbox for studying motor behavior of animals in the lab setting developed by neuroscientists at the Universities of Tübingen and Harvard. The toolbox gained significant popularity and is now used in hundreds of labs for a variety of tracking tasks. It is discussed in the corresponding Nature Neuroscience paper and was covered in the popular media such as The Harvard Gazette and The Atlantic.
Multi-person Body Pose Estimation demo at ECCV 2016
We gave a real-time demonstration of our Multi-person Body Pose Estimation algorithm at ECCV 2016.
Teaching
Probabilistic Graphical Models and their Applications
Saarland University
Winter 2015, Winter 2016Supplementary Programming Languages (Python)
Bashkir State University
Winter, 2013