13 November 2021 in IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
We address the challenging problem of whole slide image (WSI) classification. WSIs have very high resolutions and usually lack localized annotations. WSI classification can be cast as a multiple instance learning (MIL) problem when only slide-level labels are available. We propose a MIL-based method for WSI classification and tumor detection that does not require localized annotations. Our method has three major components. First, we introduce a novel MIL aggregator that models the relations of the instances in a dual-stream architecture with trainable distance measurement. Second, since WSIs can produce large or unbalanced bags that hinder the training of MIL models, we propose to use self-supervised contrastive learning to extract good representations for MIL and alleviate the issue of prohibitive memory cost for large bags. Third, we adopt a pyramidal fusion mechanism for multiscale WSI features, and further improve the accuracy of classification and localization. Our model is evaluated on two representative WSI datasets. The classification accuracy of our model compares favorably to fully-supervised methods, with less than 2% accuracy gap across datasets. Our results also outperform all previous MIL-based methods. Additional benchmark results on standard MIL datasets further demonstrate the superior performance of our MIL aggregator on general MIL problems.
9 December 2020 in Advanced Optical Materials
15 January 2021 in IEEE Transactions on Pattern Analysis and Machine Intelligence
We address the task of jointly determining what a person is doing and where they are looking based on the analysis of video captured by a headworn camera. To facilitate our research, we first introduce the EGTEA Gaze+ dataset. Our dataset comes with videos, gaze tracking data, hand masks and action annotations, thereby providing the most comprehensive benchmark for First Person Vision (FPV). Moving beyond the dataset, we propose a novel deep model for joint gaze estimation and action recognition in FPV. Our method describes the participant's gaze as a probabilistic variable and models its distribution using stochastic units in a deep network. We further sample from these stochastic units, generating an attention map to guide the aggregation of visual features for action recognition. Our method is evaluated on our EGTEA Gaze+ dataset and achieves a performance level that exceeds the state-of-the-art by a significant margin. More importantly, we demonstrate that our model can be applied to larger scale FPV dataset---EPIC-Kitchens even without using gaze, offering new state-of-the-art results on FPV action recognition.
15 September 2021 in IEEE Transactions on Antennas and Propagation
Transmissive and reflective metasurfaces have been extensively studied for generating an orbital angular momentum (OAM)-carrying beam. However, the existing OAM generators suffer from feed blockage, large profile, OAM mode nonreconfigurable, and narrowband. An OAM generator that integrates multiple functions into a single module is in high demand in OAM-based applications. In this study, we proposed an OAM generator based on a folded reflectarray with high-gain low-profile broadband and programmable characteristics. The proposed multifunctional OAM generator, comprising half the volume of the normal OAM metasurfaces, can generate a high-gain $l=0$ , $l=+1$ , and $l=+2$ OAM-mode-reconfigurable beams with high mode purity over the frequency range of 9.5–10.5 GHz. The proposed OAM generator-integrated multiple functions have potential applications in OAM-based wireless communications and radar detection.
24 June 2021 in IEEE Antennas and Wireless Propagation Letters
A planar quasi-Yagi antenna with cointegrated filtering functionality and flat in-band gain is reported. It exploits complementary-diplexer-based reflectionless-filtering techniques through the inclusion of an absorptive branch to realize the codesigned filtering functionality. This absorptive branch consists of a second-order bandstop-filter (BSF) unit loaded by a grounded resistor that is connected in parallel with the initial planar quasi-Yagi antenna. By designing the reflection coefficients of the basic antenna and the absorptive branch to be nearly complementary in frequency, the out-of-band RF power not radiated by the antenna is transmitted by the BSF to its loading resistor that dissipates it. In this manner, a quasi-absorptive filtering action is added to the antenna. Moreover, whereas the out-of-band radiation efficiency of the antenna is highly suppressed, its in-band behavior is kept as the in-band RF power is not transmitted by the BSF but radiated by the antenna as usual. For the experimental-validation purposes, a microstrip antenna prototype is developed and measured. Flat in-band gain with a ripple lower than 1 dB within the operational range of the antenna (2.6-3.06 GHz) is obtained as an added merit.