1 code implementation • CVPR 2023 • Dmitry Senushkin, Nikolay Patakin, Arseny Kuznetsov, Anton Konushin
In this work, we propose using a condition number of a linear system of gradients as a stability criterion of an MTL optimization.
no code implementations • 13 Feb 2023 • Danil Galeev, Polina Popenova, Anna Vorontsova, Anton Konushin
Recent advances in interactive segmentation (IS) allow speeding up and simplifying image editing and labeling greatly.
1 code implementation • 6 Feb 2023 • Danila Rukhovich, Anna Vorontsova, Anton Konushin
Our model with early feature fusion, which we refer to as TR3D+FF, outperforms existing 3D object detection approaches on the SUN RGB-D dataset.
Ranked #2 on
3D Object Detection
on S3DIS
(using extra training data)
1 code implementation • 6 Feb 2023 • Maksim Kolodiazhnyi, Danila Rukhovich, Anna Vorontsova, Anton Konushin
Most 3D instance segmentation methods exploit a bottom-up strategy, typically including resource-exhaustive post-processing.
Ranked #3 on
3D Instance Segmentation
on S3DIS
(using extra training data)
no code implementations • 10 Oct 2022 • Anna Sokolova, Filipp Nikitin, Anna Vorontsova, Anton Konushin
Processing large indoor scenes is a challenging task, as scan registration and camera trajectory estimation methods accumulate errors across time.
no code implementations • CVPR 2022 • Nikolay Patakin, Anna Vorontsova, Mikhail Artemyev, Anton Konushin
On the contrary, we propose GP2, General-Purpose and Geometry-Preserving training scheme, and show that conventional SVDE models can learn correct shifts themselves without any post-processing, benefiting from using stereo data even in the geometry-preserving setting.
2 code implementations • 1 Dec 2021 • Danila Rukhovich, Anna Vorontsova, Anton Konushin
Existing 3D object detection methods make prior assumptions on the geometry of objects, and we argue that it limits their generalization ability.
Ranked #5 on
3D Object Detection
on S3DIS
3 code implementations • 2 Jun 2021 • Danila Rukhovich, Anna Vorontsova, Anton Konushin
To address this problem, we propose ImVoxelNet, a novel fully convolutional method of 3D object detection based on monocular or multi-view RGB images.
Ranked #1 on
Monocular 3D Object Detection
on SUN RGB-D
6 code implementations • 12 Feb 2021 • Konstantin Sofiiuk, Ilia A. Petrov, Anton Konushin
We find that the models trained on a combination of COCO and LVIS with diverse and high-quality annotations show performance superior to all existing models.
Ranked #5 on
Interactive Segmentation
on DAVIS
no code implementations • 13 Jan 2021 • Anton Konushin, Boris Faizov, Vlad Shakhuro
Such training data is obtained by embedding synthetic images of signs in the real photos.
1 code implementation • 25 Sep 2020 • Mikhail Romanov, Nikolay Patatkin, Anna Vorontsova, Sergey Nikolenko, Anton Konushin, Dmitry Senyushkin
Our work shows that a model trained on this data along with conventional datasets can gain accuracy while predicting correct scene geometry.
1 code implementation • 18 Jun 2020 • Danil Galeev, Konstantin Sofiiuk, Danila Rukhovich, Mikhail Romanov, Olga Barinova, Anton Konushin
Based on this finding, we propose LayerMatch scheme for approximating the representation of a GAN generator that can be used for unsupervised domain-specific pretraining.
Semi-Supervised Semantic Segmentation
Vocal Bursts Intensity Prediction
1 code implementation • 1 Jun 2020 • Konstantin Sofiiuk, Polina Popenova, Anton Konushin
Image harmonization is an important step in photo editing to achieve visual consistency in composite images by adjusting the appearances of foreground to make it compatible with background.
1 code implementation • 18 May 2020 • Dmitry Senushkin, Mikhail Romanov, Ilia Belikov, Anton Konushin, Nikolay Patakin
Our second contribution is a novel training strategy that allows us to train on a semi-dense sensor data when the ground truth depth map is not available.
Ranked #1 on
Depth Completion
on Matterport3D
1 code implementation • 12 May 2020 • Danila Rukhovich, Konstantin Sofiiuk, Danil Galeev, Olga Barinova, Anton Konushin
Deep learning-based detectors usually produce a redundant set of object bounding boxes including many duplicate detections of the same object.
Ranked #1 on
Object Detection
on WiderPerson
3 code implementations • CVPR 2020 • Konstantin Sofiiuk, Ilia Petrov, Olga Barinova, Anton Konushin
We propose f-BRS (feature backpropagating refinement scheme) that solves an optimization problem with respect to auxiliary variables instead of the network inputs, and requires running forward and backward pass just for a small part of a network.
Ranked #8 on
Interactive Segmentation
on SBD
1 code implementation • 11 Dec 2019 • Igor Slinko, Anna Vorontsova, Dmitry Zhukov, Olga Barinova, Anton Konushin
We train visual odometry model on synthetic data and do not use ground truth poses hence this model can be considered unsupervised.
no code implementations • 10 Oct 2019 • David Prokhorov, Dmitry Zhukov, Olga Barinova, Anna Vorontsova, Anton Konushin
We find that while in many cases the accuracy of SLAM is very good, the robustness is still an issue.
no code implementations • 26 Sep 2019 • Pavel Kirsanov, Airat Gaskarov, Filipp Konokhov, Konstantin Sofiiuk, Anna Vorontsova, Igor Slinko, Dmitry Zhukov, Sergey Bykov, Olga Barinova, Anton Konushin
We present a novel dataset for training and benchmarking semantic SLAM methods.
1 code implementation • ICCV 2019 • Konstantin Sofiiuk, Olga Barinova, Anton Konushin
Given an input image and a point $(x, y)$, it generates a mask for the object located at $(x, y)$.
Ranked #8 on
Panoptic Segmentation
on Mapillary val
2 code implementations • 12 Sep 2019 • Nina Tuluptceva, Bart Bakker, Irina Fedulova, Anton Konushin
We present a novel method for image anomaly detection, where algorithms that use samples drawn from some distribution of "normal" data, aim to detect out-of-distribution (abnormal) samples.
no code implementations • 16 Jul 2019 • Igor Slinko, Anna Vorontsova, Filipp Konokhov, Olga Barinova, Anton Konushin
Optical Flow (OF) and depth are commonly used for visual odometry since they provide sufficient information about camera ego-motion in a rigid scene.
no code implementations • 20 Nov 2018 • Nikita Durasov, Mikhail Romanov, Valeriya Bubnova, Pavel Bogomolov, Anton Konushin
Monocular depth estimation is the task of obtaining a measure of distance for each pixel using a single image.
Indoor Monocular Depth Estimation
Monocular Depth Estimation
no code implementations • 17 Oct 2017 • Anna Sokolova, Anton Konushin
In this paper, we present a new pose-based convolutional neural network model for gait recognition.