| 1. |
- Moliner, Olivier, et al.
(författare)
-
Bootstrapped Representation Learning for Skeleton-Based Action Recognition
- 2022
-
Ingår i: Proceedings - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2022. - 2160-7508 .- 2160-7516. - 9781665487399 ; 2022-June, s. 4153-4163
-
Konferensbidrag (refereegranskat)abstract
- In this work, we study self-supervised representation learning for 3D skeleton-based action recognition. We extend Bootstrap Your Own Latent (BYOL) for representation learning on skeleton sequence data and propose a new data augmentation strategy including two asymmetric transformation pipelines. We also introduce a multi-viewpoint sampling method that leverages multiple viewing angles of the same action captured by different cameras. In the semi-supervised setting, we show that the performance can be further improved by knowledge distillation from wider networks, leveraging once more the unlabeled samples. We conduct extensive experiments on the NTU-60, NTU-120 and PKU-MMD datasets to demonstrate the performance of our proposed method. Our method consistently outperforms the current state of the art on linear evaluation, semi-supervised and transfer learning benchmarks.
|
|
| 2. |
- Valtonen Ornhag, Marcus, et al.
(författare)
-
Trust Your IMU : Consequences of Ignoring the IMU Drift
- 2022
-
Ingår i: Proceedings - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2022. - 2160-7508 .- 2160-7516. - 9781665487399 ; 2022-June, s. 4467-4476
-
Konferensbidrag (refereegranskat)abstract
- In this paper, we argue that modern pre-integration methods for inertial measurement units (IMUs) are accurate enough to ignore the drift for short time intervals. This allows us to consider a simplified camera model, which in turn admits further intrinsic calibration. We develop the first-ever solver to jointly solve the relative pose problem with unknown and equal focal length and radial distortion profile while utilizing the IMU data. Furthermore, we show significant speed-up compared to state-of-the-art algorithms, with small or negligible loss in accuracy for partially calibrated setups.The proposed algorithms are tested on both synthetic and real data, where the latter is focused on navigation using unmanned aerial vehicles (UAVs). We evaluate the proposed solvers on different commercially available low-cost UAVs, and demonstrate that the novel assumption on IMU drift is feasible in real-life applications. The extended intrinsic auto-calibration enables us to use distorted input images, making tedious calibration processes obsolete, compared to current state-of-the-art methods. Code available at: https://github.com/marcusvaltonen/DronePoseLib.1
|
|
| 3. |
- Bökman, Georg, 1994, et al.
(författare)
-
A case for using rotation invariant features in state of the art feature matchers
- 2022
-
Ingår i: IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. - 2160-7516 .- 2160-7508. ; 2022-June, s. 5106-5115
-
Konferensbidrag (refereegranskat)abstract
- The aim of this paper is to demonstrate that a state of the art feature matcher (LoFTR) can be made more robust to rotations by simply replacing the backbone CNN with a steerable CNN which is equivariant to translations and image rotations. It is experimentally shown that this boost is obtained without reducing performance on ordinary illumination and viewpoint matching sequences.
|
|
| 4. |
- Haner, Sebastian, et al.
(författare)
-
A Step Towards Self-calibration in SLAM: Weakly Calibrated On-line Structure and Motion Estimation
- 2010
-
Ingår i: Computer Vision and Pattern Recognition Workshops (CVPRW), 2010 IEEE Computer Society Conference on. - 2160-7508 .- 2160-7516. - 9781424470297 ; , s. 59-64
-
Konferensbidrag (refereegranskat)abstract
- We propose a structure and motion estimation scheme based on a dynamic systems approach, where states and parameters in a perspective system are estimated. An online method for structure and motion estimation in densely sampled image sequences is presented. The proposed method is based on an extended Kalman filter and a novel parametrization. We derive a dynamic system describing the motion of the camera and the image formation. By a change of coordinates, we represent this system by normalized image coordinates and the inverse depths. Then we apply an extended Kalman filter for estimation of both structure and motion. Furthermore, we assume only weakly calibrated cameras, i.e. cameras with unknown and possibly varying focal length, unknown and constant principal point and known aspect ratio and skew. The performance of the proposed method is demonstrated in both simulated and real experiments. We also compare our method to the one proposed by Civera et al. and show that we get superior results.
|
|
| 5. |
- Hojjat, Ali, et al.
(författare)
-
ProgDTD: Progressive Learned Image Compression with Double-Tail-Drop Training
- 2023
-
Ingår i: IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. - 2160-7516 .- 2160-7508. ; 2023-June, s. 1130-1139
-
Konferensbidrag (refereegranskat)abstract
- Progressive compression allows images to start loading as low-resolution versions, becoming clearer as more data is received. This increases user experience when, for example, network connections are slow. Today, most approaches for image compression, both classical and learned ones, are designed to be non-progressive. This paper introduces ProgDTD, a training method that transforms learned, non-progressive image compression approaches into progressive ones. The design of ProgDTD is based on the observation that the information stored within the bottleneck of a compression model commonly varies in importance. To create a progressive compression model, ProgDTD modifies the training steps to enforce the model to store the data in the bottleneck sorted by priority. We achieve progressive compression by transmitting the data in order of its sorted index. ProgDTD is designed for CNN-based learned image compression models, does not need additional parameters, and has a customizable range of progressiveness. For evaluation, we apply ProgDTD to the hyperprior model, one of the most common structures in learned image compression. Our experimental results show that ProgDTD performs comparably to its non-progressive counterparts and other state-of-the-art progressive models in terms of MS-SSIM and accuracy.
|
|
| 6. |
- Nakasi, Rose, et al.
(författare)
-
A web-based intelligence platform for diagnosis of malaria in thick blood smear images : A case for a developing country
- 2020
-
Ingår i: Proceedings - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020. - 2160-7508 .- 2160-7516. - 9781728193601 ; 2020-June, s. 4238-4244
-
Konferensbidrag (refereegranskat)abstract
- Malaria is a public health problem which affects developing countries world-wide. Inadequate skilled lab technicians in remote areas of developing countries result in untimely diagnosis of malaria parasites making it hard for effective control of the disease in highly endemic areas. The development of remote systems that can provide fast, accurate and timely diagnosis is thus a necessary innovation. With availability of internet, mobile phones and computers, rapid dissemination and timely reporting of medical image analytics is possible. This study aimed at developing and implementing an automated web-based Malaria diagnostic system for thick blood smear images under light microscopy to identify parasites. We implement an image processing algorithm based on a pre-trained model of Faster Convolutional Neural Network (Faster R-CNN) and then integrate it with web-based technology to allow easy and convenient online identification of parasites by medical practitioners. Experiments carried out on the online system with test images showed that the system could identify pathogens with a mean average precision of 0.9306. The system holds the potential to improve the efficiency and accuracy in malaria diagnosis, especially in remote areas of developing countries that lack adequate skilled labor.
|
|
| 7. |
- Ornhag, Marcus Valtonen, et al.
(författare)
-
Bilinear parameterization for differentiable rank-regularization
- 2020
-
Ingår i: IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops. - 2160-7516 .- 2160-7508. - 9781728193601 ; 2020-June, s. 1416-1425
-
Konferensbidrag (refereegranskat)abstract
- Low rank approximation is a commonly occurring problem in many computer vision and machine learning applications. There are two common ways of optimizing the resulting models. Either the set of matrices with a given rank can be explicitly parametrized using a bilinear factorization, or low rank can be implicitly enforced using regularization terms penalizing non-zero singular values. While the former approach results in differentiable problems that can be efficiently optimized using local quadratic approximation, the latter is typically not differentiable (sometimes even discontinuous) and requires first order subgradient or splitting methods. It is well known that gradient based methods exhibit slow convergence for ill-conditioned problems.In this paper we show how many non-differentiable regularization methods can be reformulated into smooth objectives using bilinear parameterization. This allows us to use standard second order methods, such as Levenberg- Marquardt (LM) and Variable Projection (VarPro), to achieve accurate solutions for ill-conditioned cases. We show on several real and synthetic experiments that our second order formulation converges to substantially more accurate solutions than competing state-of-the-art methods.1
|
|
| 8. |
- Oskarsson, Magnus
(författare)
-
Fast solvers for minimal radial distortion relative pose problems
- 2021
-
Ingår i: Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021. - 2160-7516 .- 2160-7508. - 9781665448994 ; , s. 3663-3672
-
Konferensbidrag (refereegranskat)abstract
- In this paper we present a unified formulation for a large class of relative pose problems with radial distortion and varying calibration. For minimal cases, we show that one can eliminate the number of parameters down to one to three. The relative pose can then be expressed using varying calibration constraints on the fundamental matrix, with entries that are polynomial in the parameters. We can then apply standard techniques based on the action matrix and Sturm sequences to construct our solvers. This enables efficient solvers for a large class of relative pose problems with radial distortion, using a common framework. We evaluate a number of these solvers for robust two-view inlier and epipolar geometry estimation, used as minimal solvers in RANSAC.
|
|
| 9. |
- Oskarsson, Magnus
(författare)
-
Robust image-to-image color transfer using optimal inlier maximization
- 2021
-
Ingår i: Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021. - 2160-7516 .- 2160-7508. - 9781665448994 ; , s. 786-795
-
Konferensbidrag (refereegranskat)abstract
- In this paper we target the color transfer estimation problem, when we have pixel-to-pixel correspondences. We present a feature-based method, that robustly fits color transforms to data containing gross outliers. Our solution is based on an optimal inlier maximization algorithm that maximizes the number of inliers in polynomial time. We introduce a simple feature detector and descriptor based on the structure tensor that gives the means for reliable matching of the color distributions in two images. Using combinatorial methods from optimization theory and a number of new minimal solvers, we can enumerate all possible stationary points to the inlier maximization problem. In order for our method to be tractable we use a decoupling of the intensity and color direction for a given RGB-vector. This enables the intensity transformation and the color direction transformation to be handled separately. Our method gives results comparable to state-of-the-art methods in the presence of little outliers, and large improvement for moderate or large amounts of outliers in the data. The proposed method has been tested in a number of imaging applications.
|
|
| 10. |
- Bhat, Goutam, et al.
(författare)
-
NTIRE 2022 Burst Super-Resolution Challenge
- 2022
-
Ingår i: 2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW 2022). - : IEEE. - 9781665487399 - 9781665487405 ; , s. 1040-1060
-
Konferensbidrag (refereegranskat)abstract
- Burst super-resolution has received increased attention in recent years due to its applications in mobile photography. By merging information from multiple shifted images of a scene, burst super-resolution aims to recover details which otherwise cannot be obtained using a simple input image. This paper reviews the NTIRE 2022 challenge on burst super-resolution. In the challenge, the participants were tasked with generating a clean RGB image with 4x higher resolution, given a RAW noisy burst as input. That is, the methods need to perform joint denoising, demosaicking, and super-resolution. The challenge consisted of 2 tracks. Track 1 employed synthetic data, where pixel-accurate high-resolution ground truths are available. Track 2 on the other hand used real-world bursts captured from a handheld camera, along with approximately aligned reference images captured using a DSLR. 14 teams participated in the final testing phase. The top performing methods establish a new state-of-the-art on the burst super-resolution task.
|
|
