arXiv subjects

Yifan Liu

Publications and source records attributed to Yifan Liu.

At least 181 records · Page 10Linked to original sources

arXiv2023

The state-of-the-art 3D anisotropic intracranial hemorrhage segmentation on non-contrast head CT: The INSTANCE challenge

Automatic intracranial hemorrhage segmentation in 3D non-contrast head CT (NCCT) scans is significant in clinical practice. Existing hemorrhage segmentation methods usually ignores the anisotropic nature of the NCCT, and are evaluated on different in-house datasets with distinct metrics, making it highly challenging to improve segmentation performance and perform objective comparisons among different methods. The INSTANCE 2022 was a grand challenge held in conjunction with the 2022 International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI). It is intended to resolve the above-mentioned problems and promote the development of both intracranial hemorrhage segmentation and anisotropic data processing. The INSTANCE released a training set of 100 cases with ground-truth and a validation set with 30 cases without ground-truth labels that were available to the participants. A held-out testing set with 70 cases is utilized for the final evaluation and ranking. The methods from different participants are ranked based on four metrics, including Dice Similarity Coefficient (DSC), Hausdorff Distance (HD), Relative Volume Difference (RVD) and Normalized Surface Dice (NSD). A total of 13 teams submitted distinct solutions to resolve the challenges, making several baseline models, pre-processing strategies and anisotropic data processing techniques available to future researchers. The winner method achieved an average DSC of 0.6925, demonstrating a significant growth over our proposed baseline method. To the best of our knowledge, the proposed INSTANCE challenge releases the first intracranial hemorrhage segmentation benchmark, and is also the first challenge that intended to resolve the anisotropic problem in 3D medical image segmentation, which provides new alternatives in these research fields.

eess.IV↗

arXiv2022

SegViT: Semantic Segmentation with Plain Vision Transformers

We explore the capability of plain Vision Transformers (ViTs) for semantic segmentation and propose the SegVit. Previous ViT-based segmentation networks usually learn a pixel-level representation from the output of the ViT. Differently, we make use of the fundamental component -- attention mechanism, to generate masks for semantic segmentation. Specifically, we propose the Attention-to-Mask (ATM) module, in which the similarity maps between a set of learnable class tokens and the spatial feature maps are transferred to the segmentation masks. Experiments show that our proposed SegVit using the ATM module outperforms its counterparts using the plain ViT backbone on the ADE20K dataset and achieves new state-of-the-art performance on COCO-Stuff-10K and PASCAL-Context datasets. Furthermore, to reduce the computational cost of the ViT backbone, we propose query-based down-sampling (QD) and query-based up-sampling (QU) to build a Shrunk structure. With the proposed Shrunk structure, the model can save up to $40\%$ computations while maintaining competitive performance.

cs.CV↗

arXiv2022

Prior-enhanced Temporal Action Localization using Subject-aware Spatial Attention

Temporal action localization (TAL) aims to detect the boundary and identify the class of each action instance in a long untrimmed video. Current approaches treat video frames homogeneously, and tend to give background and key objects excessive attention. This limits their sensitivity to localize action boundaries. To this end, we propose a prior-enhanced temporal action localization method (PETAL), which only takes in RGB input and incorporates action subjects as priors. This proposal leverages action subjects' information with a plug-and-play subject-aware spatial attention module (SA-SAM) to generate an aggregated and subject-prioritized representation. Experimental results on THUMOS-14 and ActivityNet-1.3 datasets demonstrate that the proposed PETAL achieves competitive performance using only RGB features, e.g., boosting mAP by 2.41% or 0.25% over the state-of-the-art approach that uses RGB features or with additional optical flow features on the THUMOS-14 dataset.

cs.CV↗

arXiv2022

Towards Accurate Reconstruction of 3D Scene Shape from A Single Monocular Image

Despite significant progress made in the past few years, challenges remain for depth estimation using a single monocular image. First, it is nontrivial to train a metric-depth prediction model that can generalize well to diverse scenes mainly due to limited training data. Thus, researchers have built large-scale relative depth datasets that are much easier to collect. However, existing relative depth estimation models often fail to recover accurate 3D scene shapes due to the unknown depth shift caused by training with the relative depth data. We tackle this problem here and attempt to estimate accurate scene shapes by training on large-scale relative depth data, and estimating the depth shift. To do so, we propose a two-stage framework that first predicts depth up to an unknown scale and shift from a single monocular image, and then exploits 3D point cloud data to predict the depth shift and the camera's focal length that allow us to recover 3D scene shapes. As the two modules are trained separately, we do not need strictly paired training data. In addition, we propose an image-level normalized regression loss and a normal-based geometry loss to improve training with relative depth annotation. We test our depth model on nine unseen datasets and achieve state-of-the-art performance on zero-shot evaluation. Code is available at: https://git.io/Depth

cs.CV↗

arXiv2022

FAKD: Feature Augmented Knowledge Distillation for Semantic Segmentation

In this work, we explore data augmentations for knowledge distillation on semantic segmentation. To avoid over-fitting to the noise in the teacher network, a large number of training examples is essential for knowledge distillation. Imagelevel argumentation techniques like flipping, translation or rotation are widely used in previous knowledge distillation framework. Inspired by the recent progress on semantic directions on feature-space, we propose to include augmentations in feature space for efficient distillation. Specifically, given a semantic direction, an infinite number of augmentations can be obtained for the student in the feature space. Furthermore, the analysis shows that those augmentations can be optimized simultaneously by minimizing an upper bound for the losses defined by augmentations. Based on the observation, a new algorithm is developed for knowledge distillation in semantic segmentation. Extensive experiments on four semantic segmentation benchmarks demonstrate that the proposed method can boost the performance of current knowledge distillation methods without any significant overhead. Code is available at: https://github.com/jianlong-yuan/FAKD.

cs.CV↗

arXiv2022

Improving Personality Consistency in Conversation by Persona Extending

Endowing chatbots with a consistent personality plays a vital role for agents to deliver human-like interactions. However, existing personalized approaches commonly generate responses in light of static predefined personas depicted with textual description, which may severely restrict the interactivity of human and the chatbot, especially when the agent needs to answer the query excluded in the predefined personas, which is so-called out-of-predefined persona problem (named OOP for simplicity). To alleviate the problem, in this paper we propose a novel retrieval-to-prediction paradigm consisting of two subcomponents, namely, (1) Persona Retrieval Model (PRM), it retrieves a persona from a global collection based on a Natural Language Inference (NLI) model, the inferred persona is consistent with the predefined personas; and (2) Posterior-scored Transformer (PS-Transformer), it adopts a persona posterior distribution that further considers the actual personas used in the ground response, maximally mitigating the gap between training and inferring. Furthermore, we present a dataset called IT-ConvAI2 that first highlights the OOP problem in personalized dialogue. Extensive experiments on both IT-ConvAI2 and ConvAI2 demonstrate that our proposed model yields considerable improvements in both automatic metrics and human evaluations.

cs.CL↗

arXiv2022

Semantic-guided Multi-Mask Image Harmonization

Previous harmonization methods focus on adjusting one inharmonious region in an image based on an input mask. They may face problems when dealing with different perturbations on different semantic regions without available input masks. To deal with the problem that one image has been pasted with several foregrounds coming from different images and needs to harmonize them towards different domain directions without any mask as input, we propose a new semantic-guided multi-mask image harmonization task. Different from the previous single-mask image harmonization task, each inharmonious image is perturbed with different methods according to the semantic segmentation masks. Two challenging benchmarks, HScene and HLIP, are constructed based on $150$ and $19$ semantic classes, respectively. Furthermore, previous baselines focus on regressing the exact value for each pixel of the harmonized images. The generated results are in the `black box' and cannot be edited. In this work, we propose a novel way to edit the inharmonious images by predicting a series of operator masks. The masks indicate the level and the position to apply a certain image editing operation, which could be the brightness, the saturation, and the color in a specific dimension. The operator masks provide more flexibility for users to edit the image further. Extensive experiments verify that the operator mask-based network can further improve those state-of-the-art methods which directly regress RGB images when the perturbations are structural. Experiments have been conducted on our constructed benchmarks to verify that our proposed operator mask-based framework can locate and modify the inharmonious regions in more complex scenes. Our code and models are available at https://github.com/XuqianRen/Semantic-guided-Multi-mask-Image-Harmonization.git.

cs.CV↗

arXiv2022

Controllable Shadow Generation Using Pixel Height Maps

Shadows are essential for realistic image compositing. Physics-based shadow rendering methods require 3D geometries, which are not always available. Deep learning-based shadow synthesis methods learn a mapping from the light information to an object's shadow without explicitly modeling the shadow geometry. Still, they lack control and are prone to visual artifacts. We introduce pixel heigh, a novel geometry representation that encodes the correlations between objects, ground, and camera pose. The pixel height can be calculated from 3D geometries, manually annotated on 2D images, and can also be predicted from a single-view RGB image by a supervised approach. It can be used to calculate hard shadows in a 2D image based on the projective geometry, providing precise control of the shadows' direction and shape. Furthermore, we propose a data-driven soft shadow generator to apply softness to a hard shadow based on a softness input parameter. Qualitative and quantitative evaluations demonstrate that the proposed pixel height significantly improves the quality of the shadow generation while allowing for controllability.

cs.CV↗

arXiv2022

Ultra-sensitive Flexible Sponge-Sensor Array for Muscle Activities Detection and Human Limb Motion Recognition

Human limb motion tracking and recognition plays an important role in medical rehabilitation training, lower limb assistance, prosthetics design for amputees, feedback control for assistive robots, etc. Lightweight wearable sensors, including inertial sensors, surface electromyography sensors, and flexible strain/pressure, are promising to become the next-generation human motion capture devices. Herein, we present a wireless wearable device consisting of a sixteen-channel flexible sponge-based pressure sensor array to recognize various human lower limb motions by detecting contours on the human skin caused by calf gastrocnemius muscle actions. Each sensing element is a round porous structure of thin carbon nanotube/polydimethylsiloxane nanocomposites with a diameter of 4 mm and thickness of about 400 μm. Ten human subjects were recruited to perform ten different lower limb motions while wearing the developed device. The motion classification result with the support vector machine method shows a macro-recall of about 97.3% for all ten motions tested. This work demonstrates a portable wearable muscle activity detection device with a lower limb motion recognition application, which can be potentially used in assistive robot control, healthcare, sports monitoring, etc.

eess.SP↗

arXiv2022

$B^p_r(F_n)$ has no nontrivial idempotents

We show that there is no nontrivial idempotent in the reduced group $\ell^p$-operator algebra $B^p_r(F_n)$ of the free group $F_n$ on $n$ generators for each positive integer $n$.

math.OA↗

arXiv2022

Passive Photoacoustic Effect

Photoacoustic effect refers to the acoustic generation induced by laser irradiation, where nanosecond pulsed laser source is normally used to provide instantaneous heating and thermoelastic expansion of the sample. More generally, photoacoustic generation requires active intensity modulation of laser source to produce sharp temperature gradient, which is key to generate acoustic wave. In this paper, we propose a novel photoacoustic effect from moving droplet, which generates photoacoustic wave by a non-modulated continuous-wave laser. When the droplet moves through the laser spot area, it can be heated up and cooled down instantaneously, generating photoacoustic waves. We name it passive photoacoustic effect. Theoretical analysis and simulation study validated the existence of passive photoacoustic effect. This phenomenon may find potential application in high-throughput photoacoustic cytometry.

physics.app-ph↗

arXiv2022

Direction and Trajectory Tracking Control for Nonholonomic Spherical Robot by Combining Sliding Mode Controller and Model Prediction Controller

Spherical robot is a nonlinear, nonholonomic and unstable system which increases the difficulty of the direction and trajectory tracking problem. In this study, we propose a new direction controller HTSMC, an instruction planning controller MPC, and a trajectory tracking framework MHH. The HTSMC is designed by integrating a fast terminal algorithm, a hierarchical method, the motion features of a spherical robot, and its dynamics. In addition, the new direction controller has an excellent control effect with a quick response speed and strong stability. MPC can obtain optimal commands that are then transmitted to the velocity and direction controller. Since the two torque controllers in MHH are all Lyapunov-based sliding mode controllers, the MHH framework may achieve optimal control performance while assuring stability. Finally, the two controllers eliminate the requirement for MPC's stability and dynamic constraints. Finally, hardware experiments demonstrate the efficacy of the HTSMC, MPC, and MHH.

cs.RO↗

arXiv2022

Elevation Angle-Dependent 3D Trajectory Design for Aerial RIS-aided Communication

This paper investigates an aerial reconfigurable intelligent surface (RIS)-aided communication system under the probabilistic line-of-sight (LoS) channel, where an unmanned aerial vehicle (UAV) equipped with an RIS is deployed to assist two ground nodes in their information exchange. An optimization problem with the objective of maximizing the minimum average achievable rate is formulated to jointly design the communication scheduling, the RIS's phase shift, and the three-dimensional (3D) UAV trajectory. To solve such a non-convex problem, we propose an efficient iterative algorithm to obtain its suboptimal solution. Simulation results show that our proposed design significantly outperforms the existing schemes and provides new insights into the elevation angle and distance trade-off for the UAV-borne RIS communication system.

eess.SP↗

arXiv2022

Efficient Video Segmentation Models with Per-frame Inference

Most existing real-time deep models trained with each frame independently may produce inconsistent results across the temporal axis when tested on a video sequence. A few methods take the correlations in the video sequence into account,e.g., by propagating the results to the neighboring frames using optical flow or extracting frame representations using multi-frame information, which may lead to inaccurate results or unbalanced latency. In this work, we focus on improving the temporal consistency without introducing computation overhead in inference. To this end, we perform inference at each frame. Temporal consistency is achieved by learning from video frames with extra constraints during the training phase. introduced for inference. We propose several techniques to learn from the video sequence, including a temporal consistency loss and online/offline knowledge distillation methods. On the task of semantic video segmentation, weighing among accuracy, temporal smoothness, and efficiency, our proposed method outperforms keyframe-based methods and a few baseline methods that are trained with each frame independently, on datasets including Cityscapes, Camvid, and 300VW-Mask. We further apply our training method to video instance segmentation on YouTubeVISand develop an application of portrait matting in video sequences, by segmenting temporally consistent instance-level trimaps across frames. Experiments show superior qualitative and quantitative results. Code is available at: https://git.io/vidseg.

cs.CV↗

arXiv2022

Full-Dimensional Rate Enhancement for UAV-Enabled Communications via Intelligent Omni-Surface

This paper investigates the achievable rate maximization problem of a downlink unmanned aerial vehicle (UAV)-enabled communication system aided by an intelligent omni-surface (IOS). Different from the state-of-the-art reconfigurable intelligent surface (RIS) that only reflects incident signals, the IOS can simultaneously reflect and transmit the signals, thereby providing full-dimensional rate enhancement. To tackle such a problem, we formulate it by jointly optimizing the IOS's phase shift and the UAV trajectory. Although it is difficult to solve it optimally due to its non-convexity, we propose an efficient iterative algorithm to obtain a high-quality suboptimal solution. Simulation results show that the IOS-assisted UAV communications can achieve more significant improvement in achievable rates than other benchmark schemes.

eess.SY↗

arXiv2022

Learning-based multiplexed transmission of scattered twisted light through a kilometer-scale standard multimode fiber

Multiplexing multiple orbital angular momentum (OAM) modes of light has the potential to increase data capacity in optical communication. However, the distribution of such modes over long distances remains challenging. Free-space transmission is strongly influenced by atmospheric turbulence and light scattering, while the wave distortion induced by the mode dispersion in fibers disables OAM demultiplexing in fiber-optic communications. Here, a deep-learning-based approach is developed to recover the data from scattered OAM channels without measuring any phase information. Over a 1-km-long standard multimode fiber, the method is able to identify different OAM modes with an accuracy of more than 99.9% in parallel demultiplexing of 24 scattered OAM channels. To demonstrate the transmission quality, color images are encoded in multiplexed twisted light and our method achieves decoding the transmitted data with an error rate of 0.13%. Our work shows the artificial intelligence algorithm could benefit the use of OAM multiplexing in commercial fiber networks and high-performance optical communication in turbulent environments.

physics.optics↗

arXiv2021

X-GGM: Graph Generative Modeling for Out-of-Distribution Generalization in Visual Question Answering

Encouraging progress has been made towards Visual Question Answering (VQA) in recent years, but it is still challenging to enable VQA models to adaptively generalize to out-of-distribution (OOD) samples. Intuitively, recompositions of existing visual concepts (\ie, attributes and objects) can generate unseen compositions in the training set, which will promote VQA models to generalize to OOD samples. In this paper, we formulate OOD generalization in VQA as a compositional generalization problem and propose a graph generative modeling-based training scheme (X-GGM) to implicitly model the problem. X-GGM leverages graph generative modeling to iteratively generate a relation matrix and node representations for the predefined graph that utilizes attribute-object pairs as nodes. Furthermore, to alleviate the unstable training issue in graph generative modeling, we propose a gradient distribution consistency loss to constrain the data distribution with adversarial perturbations and the generated distribution. The baseline VQA model (LXMERT) trained with the X-GGM scheme achieves state-of-the-art OOD performance on two standard VQA OOD benchmarks, \ie, VQA-CP v2 and GQA-OOD. Extensive ablation studies demonstrate the effectiveness of X-GGM components. Code is available at \url{https://github.com/jingjing12110/x-ggm}.

cs.CV↗

arXiv2021

Channel-wise Knowledge Distillation for Dense Prediction

Knowledge distillation (KD) has been proven to be a simple and effective tool for training compact models. Almost all KD variants for dense prediction tasks align the student and teacher networks' feature maps in the spatial domain, typically by minimizing point-wise and/or pair-wise discrepancy. Observing that in semantic segmentation, some layers' feature activations of each channel tend to encode saliency of scene categories (analogue to class activation mapping), we propose to align features channel-wise between the student and teacher networks. To this end, we first transform the feature map of each channel into a probabilty map using softmax normalization, and then minimize the Kullback-Leibler (KL) divergence of the corresponding channels of the two networks. By doing so, our method focuses on mimicking the soft distributions of channels between networks. In particular, the KL divergence enables learning to pay more attention to the most salient regions of the channel-wise maps, presumably corresponding to the most useful signals for semantic segmentation. Experiments demonstrate that our channel-wise distillation outperforms almost all existing spatial distillation methods for semantic segmentation considerably, and requires less computational cost during training. We consistently achieve superior performance on three benchmarks with various network structures. Code is available at: https://git.io/Distiller

cs.CV↗