| 1. |
- Kristan, Matej, et al.
(författare)
-
The first visual object tracking segmentation VOTS2023 challenge results
- 2023
-
Ingår i: 2023 IEEE/CVF International conference on computer vision workshops (ICCVW). - : Institute of Electrical and Electronics Engineers Inc.. - 9798350307443 - 9798350307450 ; , s. 1788-1810
-
Konferensbidrag (refereegranskat)abstract
- The Visual Object Tracking Segmentation VOTS2023 challenge is the eleventh annual tracker benchmarking activity of the VOT initiative. This challenge is the first to merge short-term and long-term as well as single-target and multiple-target tracking with segmentation masks as the only target location specification. A new dataset was created; the ground truth has been withheld to prevent overfitting. New performance measures and evaluation protocols have been created along with a new toolkit and an evaluation server. Results of the presented 47 trackers indicate that modern tracking frameworks are well-suited to deal with convergence of short-term and long-term tracking and that multiple and single target tracking can be considered a single problem. A leaderboard, with participating trackers details, the source code, the datasets, and the evaluation kit are publicly available at the challenge website1
|
|
| 2. |
- Zhu, Lingxiao, et al.
(författare)
-
A NoC-Based Spatial DNN Inference Accelerator With Memory-Friendly Dataflow
- 2023
-
Ingår i: IEEE design & test. - : Institute of Electrical and Electronics Engineers (IEEE). - 2168-2356 .- 2168-2364. ; 40:6, s. 39-50
-
Tidskriftsartikel (refereegranskat)abstract
- This article addresses the challenges of excessive storage overhead and the absence of sparsity-aware design in Network-on-Chip (NoC)-based spatial deep neural network accelerators. The authors present a prototype chip that outperforms existing accelerators in both energy and area efficiency, demonstrated on TSMC 28-nm process technology. —Mahdi Nikdast, Colorado State University, USA —Miquel Moreto, Barcelona Supercomputing Center, Spain —Masoumeh (Azin) Ebrahimi, KTH Royal Institute of Technology, Sweden —Sujay Deb, IIIT Delhi, India
|
|
| 3. |
- Boudreau, Mathieu, et al.
(författare)
-
Repeat it without me: Crowdsourcing the T1 mapping common ground via the ISMRM reproducibility challenge
- 2024
-
Ingår i: MAGNETIC RESONANCE IN MEDICINE. - 0740-3194 .- 1522-2594. ; 92:3, s. 1115-1127
-
Tidskriftsartikel (refereegranskat)abstract
- Purpose T-1 mapping is a widely used quantitative MRI technique, but its tissue-specific values remain inconsistent across protocols, sites, and vendors. The ISMRM Reproducible Research and Quantitative MR study groups jointly launched a challenge to assess the reproducibility of a well-established inversion-recovery T-1 mapping technique, using acquisition details from a seminal T-1 mapping paper on a standardized phantom and in human brains. Methods The challenge used the acquisition protocol from Barral et al. (2010). Researchers collected T-1 mapping data on the ISMRM/NIST phantom and/or in human brains. Data submission, pipeline development, and analysis were conducted using open-source platforms. Intersubmission and intrasubmission comparisons were performed. Results Eighteen submissions (39 phantom and 56 human datasets) on scanners by three MRI vendors were collected at 3 T (except one, at 0.35 T). The mean coefficient of variation was 6.1% for intersubmission phantom measurements, and 2.9% for intrasubmission measurements. For humans, the intersubmission/intrasubmission coefficient of variation was 5.9/3.2% in the genu and 16/6.9% in the cortex. An interactive dashboard for data visualization was also eveloped: https://rrsg2020.dashboards.neurolibre.org. Conclusion The T-1 intersubmission variability was twice as high as the intrasubmission variability in both phantoms and human brains, indicating that the acquisition details in the original paper were insufficient to reproduce a quantitative MRI protocol. This study reports the inherent uncertainty in T-1 measures across independent research groups, bringing us one step closer to a practical clinical baseline of T-1 variations in vivo.
|
|
| 4. |
- Chen, Zemeng, et al.
(författare)
-
Open access dataset integrating EEG and fNIRS during Stroop tasks
- 2023
-
Ingår i: Scientific Data. - : Springer Nature. - 2052-4463. ; 10:1
-
Tidskriftsartikel (refereegranskat)abstract
- Conflict monitoring and processing are crucial components of the human cognitive system, with significant implications for daily life and the diagnosis of cognitive disorders. The Stroop task, combined with brain function detection technology, has been widely employed as a classical paradigm for investigating conflict processing. However, there remains a lack of public datasets that integrate Electroencephalogram (EEG) and functional Near-infrared Spectroscopy (fNIRS) to simultaneously record brain activity during a Stroop task. We introduce a dual-modality Stroop task dataset incorporating 34-channel EEG (sampling frequency is 1000 Hz) and 20-channel high temporal resolution fNIRS (sampling frequency is 100 Hz) measurements covering the whole frontal cerebral cortex from 21 participants (9 females/12 males, aged 23.0 ± 2.3 years). Event-related potential analysis of EEG recordings and activation analysis of fNIRS recordings were performed to show the significant Stroop effect. We expected that the data provided would be utilized to investigate multimodal data processing algorithms during cognitive processing.
|
|
| 5. |
- Li, Chenyang, et al.
(författare)
-
Dynamically weighted multireference perturbation theory : Combining the advantages of multi-state and state-averaged methods
- 2019
-
Ingår i: Journal of Chemical Physics. - : AIP Publishing. - 0021-9606 .- 1089-7690. ; 150:14
-
Tidskriftsartikel (refereegranskat)abstract
- We introduce two new approaches to compute near-degenerate electronic states based on the driven similarity renormalization group (DSRG) framework. The first approach is a unitary multi-state formalism based on the DSRG (MS-DSRG), whereby an effective Hamiltonian is built from a set of state-specific solutions. The second approach employs a dynamic weighting parameter to smoothly interpolate between the multi-state and the state-averaged DSRG schemes. The resulting dynamically weighted DSRG (DW-DSRG) theory incorporates the most desirable features of both multi-state approaches (ability to accurately treat many states) and state-averaged methods (correct description of avoided crossings and conical intersections). We formulate second-order perturbation theories (PT2) based on the MS-and DW-DSRG and study the potential energy curves of LiF, the conical intersection of the two lowest singlet states of NH3, and several low-lying excited states of benzene, naphthalene, and anthracene. The DW-DSRG-PT2 predicts the correct avoided crossing of LiF and avoids artifacts produced by the corresponding state-specific and multi-state theories. Excitation energies of the acenes computed with the DW-DSRG-PT2 are found to be more accurate than the corresponding state-averaged values, showing a small dependence on the number of states computed.
|
|
| 6. |
- Li, Huafu, et al.
(författare)
-
Performance Analysis and Transmission Block Size Optimization for Massive MIMO Vehicular Network with Spatially and Temporally Correlated Channels
- 2024
-
Ingår i: IEEE Internet of Things Journal. - 2327-4662. ; 11:5, s. 8989-9003
-
Tidskriftsartikel (refereegranskat)abstract
- We investigate the effect of spatially and temporally correlated channels on the transmission performance of multi-cell multi-user massive multiple-input multiple-output (MIMO) vehicular networks in generic non-isotropic scattering environments. A new channel model is established to evaluate the harmfulness of the non-isotropic-scattered angle-of-departure/angle-of-arrival (AoD/AoA) spread and the high mobility of users on the uplink transmission. We derive the expressions of achievable spectral efficiency (SE), taking into account the effects of line-of-sight propagation, channel aging, and pilot contamination. Specifically, two novel receive combining schemes, namely the aging-aware maximum ratio combining and the aging-aware minimum mean square error combining, are presented to mitigate the SE decline caused by outdated channel state information. A low-complexity pilot assignment algorithm is proposed to suppress pilot contamination. We find that the quasi-static assumption of the channel may be unsafe for the system design of the vehicular networks even within a single transmission block period lasting from hundreds of microseconds to a few milliseconds. We observe that there exists an optimal block size Copt that maximizes area spectral efficiency. Especially, Copt can be expressed as a function of movement speed, AoD spread, and AoA spread. Numerical results are presented to validate the efficacy of the proposed schemes and highlight the importance of correct performance evaluation for practical massive MIMO system designs.
|
|
