Jianzhuang Liu's research while affiliated with Shenzhen Institute of Standards and Technology and other places
What is this page?
This page lists the scientific contributions of an author, who either does not have a ResearchGate profile, or has not yet added these contributions to their profile.
It was automatically created by ResearchGate to create a record of this author's body of work. We create such pages to advance our goal of creating and maintaining the most comprehensive scientific repository possible. In doing so, we process publicly available (personal) data relating to the author as a member of the scientific community.
If you're a ResearchGate member, you can follow this page to keep up with this author's work.
If you are this author, and you don't want us to display this page anymore, please let us know.
It was automatically created by ResearchGate to create a record of this author's body of work. We create such pages to advance our goal of creating and maintaining the most comprehensive scientific repository possible. In doing so, we process publicly available (personal) data relating to the author as a member of the scientific community.
If you're a ResearchGate member, you can follow this page to keep up with this author's work.
If you are this author, and you don't want us to display this page anymore, please let us know.
Publications (281)
Vision-Language Navigation (VLN) requires the agent to follow language instructions to reach a target position. A key factor for successful navigation is to align the landmarks implied in the instruction with diverse visual observations. However, previous VLN agents fail to perform accurate modality alignment especially in unexplored scenes, since...
Vision-Language Navigation (VLN) requires the agent to follow language instructions to reach a target position. A key factor for successful navigation is to align the landmarks implied in the instruction with diverse visual observations. However, previous VLN agents fail to perform accurate modality alignment especially in unexplored scenes, since...
3D Gaussian Splatting showcases notable advancements in photo-realistic and real-time novel view synthesis. However, it faces challenges in modeling mirror reflections, which exhibit substantial appearance variations from different viewpoints. To tackle this problem, we present MirrorGaussian, the first method for mirror scene reconstruction with r...
Brain signal visualization has emerged as an active research area, serving as a critical interface between the human visual system and computer vision models. Diffusion-based methods have recently shown promise in analyzing functional magnetic resonance imaging (fMRI) data, including the reconstruction of high-quality images consistent with origina...
Self-supervised learning aims to learn representation that can be effectively generalized to downstream tasks. Many self-supervised approaches regard two views of an image as both the input and the self-supervised signals, assuming that either view contains the same task-relevant information and the shared information is (approximately) sufficient...
The stereo event-intensity camera setup is widely applied to leverage the advantages of both event cameras with low latency and intensity cameras that capture accurate brightness and texture information. However, such a setup commonly encounters cross-modality parallax that is difficult to be eliminated solely with stereo rectification especially f...
Facial manipulation techniques have aroused increasing security concerns, leading to various methods to detect forgery videos. However, existing methods suffer from a significant performance gap compared to image manipulation methods, partially because the spatio-temporal information is not well explored. To address the issue, we introduce a Hybrid...
This paper presents a novel network structure with illumination-aware gamma correction and complete image modelling to solve the low-light image enhancement problem. Low-light environments usually lead to less informative large-scale dark areas, directly learning deep representations from low-light images is insensitive to recovering normal illumin...
Event-based motion deblurring has shown promising results by exploiting low-latency events. However, current approaches are limited in their practical usage, as they assume the same spatial resolution of inputs and specific blurriness distributions. This work addresses these limitations and aims to generalize the performance of event-based deblurri...
Recently, semantic segmentation models trained with image-level text supervision have shown promising results in challenging open-world scenarios. However, these models still face difficulties in learning fine-grained semantic alignment at the pixel level and predicting accurate object masks. To address this issue, we propose MixReorg, a novel and...
The stereo event-intensity camera setup is widely applied to leverage the advantages of both event cameras with low latency and intensity cameras that capture accurate brightness and texture information. However, such a setup commonly encounters cross-modality parallax that is difficult to be eliminated solely with stereo rectification especially f...
Low-light video enhancement (LLVE) is an important yet challenging task with many applications such as photographing and autonomous driving. Unlike single image low-light enhancement, most LLVE methods utilize temporal information from adjacent frames to restore the color and remove the noise of the target frame. However, these algorithms, based on...
Vision-Language Navigation (VLN) is a challenging task which requires an agent to align complex visual observations to language instructions to reach the goal position. Most existing VLN agents directly learn to align the raw directional features and visual features trained using one-hot labels to linguistic instruction features. However, the big s...
Continual learning aims to enable a model to incrementally learn knowledge from sequentially arrived data. Previous works adopt the conventional classification architecture, which consists of a feature extractor and a classifier. The feature extractor is shared across sequentially arrived tasks or classes, but one specific group of weights of the c...
Pre-trained vision-language models have inspired much research on few-shot learning. However, with only a few training images, there exist two crucial problems: (1) the visual feature distributions are easily distracted by class-irrelevant information in images, and (2) the alignment between the visual and language feature distributions is difficul...
Despite the substantial progress of active learning for image recognition, there lacks a systematic investigation of instance-level active learning for object detection. In this paper, we propose to unify instance uncertainty calculation with image uncertainty estimation for informative image selection, creating a multiple instance differentiation...
Brain signal visualization has emerged as an active research area, serving as a critical interface between the human visual system and computer vision models. Although diffusion models have shown promise in analyzing functional magnetic resonance imaging (fMRI) data, including reconstructing high-quality images consistent with original visual stimu...
In recent years, videos and images in 720p (HD), 1080p (FHD) and 4K (UHD) resolution have become more popular for display devices such as TVs, mobile phones and VR. However, these high resolution images cannot achieve the expected visual effect due to the limitation of the internet bandwidth, and bring a great challenge for super-resolution network...
Medical artificial general intelligence (MAGI) enables one foundation model to solve different medical tasks, which is very practical in the medical domain. It can significantly reduce the requirement of large amounts of task-specific data by sufficiently sharing medical knowledge among different tasks. However, due to the challenges of designing s...
Face animation has achieved much progress in computer vision. However, prevailing GAN-based methods suffer from unnatural distortions and artifacts due to sophisticated motion deformation. In this paper, we propose a Face Animation framework with an attribute-guided Diffusion Model (FADM), which is the first work to exploit the superior modeling ca...
This paper aims at demystifying a single motion-blurred image with events and revealing temporally continuous scene dynamics encrypted behind motion blurs. To achieve this end, an Implicit Video Function (IVF) is learned to represent a single motion blurred image with concurrent events, enabling the latent sharp image restoration of arbitrary times...
![Quantitative comparison (PSNR and LPIPS) on LSUN [47] with 16× SR.](profile/Sheng-Xu-54/publication/369624026/figure/tbl2/AS:11431281131561937@1680146307860/Quantitative-comparison-PSNR-and-LPIPS-on-LSUN-47-with-16-SR_Q320.jpg)
Image super-resolution (SR) has attracted increasing attention due to its wide applications. However, current SR methods generally suffer from over-smoothing and artifacts, and most work only with fixed magnifications. This paper introduces an Implicit Diffusion Model (IDM) for high-fidelity continuous image super-resolution. IDM integrates an impl...
Super-Resolution from a single motion Blurred image (SRB) is a severely ill-posed problem due to the joint degradation of motion blurs and low spatial resolution. In this paper, we employ events to alleviate the burden of SRB and propose an Event-enhanced SRB (E-SRB) algorithm, which can generate a sequence of sharp and clear images with High Resol...
italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S
uper-
R
esolution from a single motion
B
lurred image (SRB) is a severely ill-posed problem due to the joint degradation of motion blurs and low spatial resolution. In this paper, we employ events to alleviate the burden of SRB and propose an
E...
Recently, great success has been made in learning visual representations from text supervision, facilitating the emergence of text-supervised semantic segmentation. However, existing works focus on pixel grouping and cross-modal semantic alignment, while ignoring the correspondence among multiple augmented views of the same image. To overcome such...
Pedestrian detection in the wild remains a challenging problem especially for scenes containing serious occlusion. In this paper, we propose a novel feature learning method in the deep learning framework, referred to as Feature Calibration Network (FC-Net), to adaptively detect pedestrians under various occlusions. FC-Net is based on the observatio...
Referring image segmentation aims at localizing all pixels of the visual objects described by a natural language sentence. Previous works learn to straightforwardly align the sentence embedding and pixel-level embedding for highlighting the referred objects, but ignore the semantic consistency of pixels within the same object, leading to incomplete...
Existing few-shot learning (FSL) methods rely on training with a large labeled dataset, which prevents them from leveraging abundant unlabeled data. From an information-theoretic perspective, we propose an effective unsupervised FSL method, learning representations with self-supervision. Following the InfoMax principle, our method learns comprehens...
Humans can continuously learn new knowledge. However, machine learning models suffer from drastic dropping in performance on previous tasks after learning new tasks. Cognitive science points out that the competition of similar knowledge is an important cause of forgetting. In this paper, we design a paradigm for lifelong learning based on meta-lear...
Top-down methods dominate the field of 3D human pose and shape estimation, because they are decoupled from human detection and allow researchers to focus on the core problem. However, cropping, their first step, discards the location information from the very beginning, which makes themselves unable to accurately predict the global rotation in the...
Face animation, one of the hottest topics in computer vision, has achieved a promising performance with the help of generative models. However, it remains a critical challenge to generate identity preserving and photo-realistic images due to the sophisticated motion deformation and complex facial detail modeling. To address these problems, we propo...
Though graph representation learning (GRL) has made significant progress, it is still a challenge to extract and embed the rich topological structure and feature information in an adequate way. Most existing methods focus on local structure and fail to fully incorporate the global topological structure. To this end, we propose a novel Structure-Pre...
Humans can continuously learn new knowledge. However, machine learning models suffer from drastic dropping in performance on previous tasks after learning new tasks. Cognitive science points out that the competition of similar knowledge is an important cause of forgetting. In this paper, we design a paradigm for lifelong learning based on meta-lear...
Due to the difficulty in collecting paired real-world training data, image deraining is currently dominated by supervised learning with synthesized data generated by e.g., Photoshop rendering. However, the generalization to real rainy scenes is usually limited due to the gap between synthetic and real-world data. In this paper, we first statistical...
Low-light video enhancement (LLVE) is an important yet challenging task with many applications such as photographing and autonomous driving. Unlike single image low-light enhancement, most LLVE methods utilize temporal information from adjacent frames to restore the color and remove the noise of the target frame. However, these algorithms, based on...
Top-down methods dominate the field of 3D human pose and shape estimation, because they are decoupled from human detection and allow researchers to focus on the core problem. However, cropping, their first step, discards the location information from the very beginning, which makes themselves unable to accurately predict the global rotation in the...
Existing few-shot learning (FSL) methods rely on training with a large labeled dataset, which prevents them from leveraging abundant unlabeled data. From an information-theoretic perspective, we propose an effective unsupervised FSL method, learning representations with self-supervision. Following the InfoMax principle, our method learns comprehens...
Deep learning has made remarkable achievements for single image haze removal. However, existing deep dehazing models only give deterministic results without discussing the uncertainty of them. There exist two types of uncertainty in the dehazing models: aleatoric uncertainty that comes from noise inherent in the observations and epistemic uncertain...
We propose a novel zero-shot multi-frame image restoration method for removing unwanted obstruction elements (such as rains, snow, and moire patterns) that vary in successive frames. It has three stages: transformer pre-training, zero-shot restoration, and hard patch refinement. Using the pre-trained transformers, our model is able to tell the moti...
In recent years, image denoising has benefited a lot from deep neural networks. However, these models need large amounts of noisy-clean image pairs for supervision. Although there have been attempts in training denoising networks with only noisy images, existing self-supervised algorithms suffer from inefficient network training, heavy computationa...
Vision-Language Navigation (VLN) is a challenging task that requires an embodied agent to perform action-level modality alignment, i.e., make instruction-asked actions sequentially in complex visual environments. Most existing VLN agents learn the instruction-path data directly and cannot sufficiently explore action-level alignment knowledge inside...
We present prompt distribution learning for effectively adapting a pre-trained vision-language model to address downstream recognition tasks. Our method not only learns low-bias prompts from a few samples but also captures the distribution of diverse prompts to handle the varying visual representations. In this way, we provide high-quality task-rel...
Color constancy aims to restore the constant colors of a scene under different illuminants. However, due to the existence of camera spectral sensitivity, the network trained on a certain sensor, cannot work well on others. Also, since the training datasets are collected in certain environments, the diversity of illuminants is limited for complex re...
Referring expression comprehension (REC) aims to locate a certain object in an image referred by a natural language expression. For joint understanding of regions and expressions, existing REC works typically target on modeling the cross-modal relevance in each region-expression pair within each single image. In this paper, we explore a new but gen...
Self-attention (SA) based networks have achieved great success in image captioning, constantly dominating the leaderboards of online benchmarks. However, existing SA networks still suffer from distance insensitivity and low-rank bottleneck. In this paper, we aim to optimize SA in terms of two aspects, thereby addressing the above issues. First, we...
We propose a novel zero-shot multi-frame image restoration method for removing unwanted obstruction elements (such as rains, snow, and moire patterns) that vary in successive frames. It has three stages: transformer pre-training, zero-shot restoration, and hard patch refinement. Using the pre-trained transformers, our model is able to tell the moti...
Learning to synthesize data has emerged as a promising direction in zero-shot quantization (ZSQ), which represents neural networks by low-bit integer without accessing any of the real data. In this paper, we observe an interesting phenomenon of intra-class heterogeneity in real data and show that existing methods fail to retain this property in the...
In this paper, we propose an end-to-end learning framework for event-based motion deblurring in a self-supervised manner, where real-world events are exploited to alleviate the performance degradation caused by data inconsistency. To achieve this end, optical flows are predicted from events, with which the blurry consistency and photometric consist...
A resource-adaptive supernet adjusts its subnets for inference to fit the dynamically available resources. In this paper, we propose Prioritized Subnet Sampling to train a resource-adaptive supernet, termed PSS-Net. We maintain multiple subnet pools, each of which stores the information of substantial subnets with similar resource consumption. Cons...
High dynamic range (HDR) imaging from multiple low dynamic range (LDR) images has been suffering from ghosting artifacts caused by scene and objects motion. Existing methods, such as optical flow based and end-to-end deep learning based solutions, are error-prone either in detail restoration or ghosting artifacts removal. Comprehensive empirical ev...
The mainstream approach for filter pruning is usually either to force a hard-coded importance estimation upon a computation-heavy pretrained model to select "important" filters, or to impose a hyperparameter-sensitive sparse constraint on the loss objective to regularize the network training. In this paper, we present a novel filter pruning method,...
We propose a novel network pruning approach by information preserving of pretrained network weights (filters). Network pruning with the information preserving is formulated as a matrix sketch problem, which is efficiently solved by the off-the-shelf frequent direction method. Our approach, referred to as FilterSketch, encodes the second-order infor...
Recently unsupervised domain adaptation for the semantic segmentation task has become more and more popular due to high-cost of pixel-level annotation on real-world images. However, most domain adaptation methods are only restricted to single-source-single-target pair, and can not be directly extended to multiple target domains. In this work, we pr...
In this paper, we present Uformer, an effective and efficient Transformer-based architecture, in which we build a hierarchical encoder-decoder network using the Transformer block for image restoration. Uformer has two core designs to make it suitable for this task. The first key element is a local-enhanced window Transformer block, where we use non...
Channel pruning and tensor decomposition have received extensive attention in convolutional neural network compression. However, these two techniques are traditionally deployed in an isolated manner, leading to significant accuracy drop when pursuing high compression rates. In this paper, we propose a Collaborative Compression (CC) scheme, which jo...
In this paper, we propose a domain-general model, termed learning-to-weight (LTW), that guarantees face detection performance across multiple domains, particularly the target domains that are never seen before. However, various face forgery methods cause complex and biased data distributions, making it challenging to detect fake faces in unseen dom...
Domain generalization (DG) offers a preferable real-world setting for Person Re-Identification (Re-ID), which trains a model using multiple source domain datasets and expects it to perform well in an unseen target domain without any model updating. Unfortunately, most DG approaches are designed explicitly for classification tasks, which fundamental...
Despite the substantial progress of active learning for image recognition, there still lacks an instance-level active learning method specified for object detection. In this paper, we propose Multiple Instance Active Object Detection (MI-AOD), to select the most informative images for detector training by observing instance-level uncertainty. MI-AO...
Binarized convolutional neural networks (BNNs) are widely used to improve the memory and computational efficiency of deep convolutional neural networks for to be employed on embedded devices. However, existing BNNs fail to explore their corresponding full-precision models’ potential, resulting in a significant performance gap. This paper introduces...
Binary neural networks (BNNs) have received increasing attention due to their superior reductions of computation and memory. Most existing works focus on either lessening the quantization error by minimizing the gap between the full-precision weights and their binarization or designing a gradient approximation to mitigate the gradient mismatch, whi...
This paper presents a learning-based approach to synthesize the view from an arbitrary camera position given a sparse set of images. A key challenge for this novel view synthesis arises from the reconstruction process, when the views from different input images may not be consistent due to obstruction in the light path. We overcome this by jointly...
Multi-source unsupervised domain adaptation~(MSDA) aims at adapting models trained on multiple labeled source domains to an unlabeled target domain. In this paper, we propose a novel multi-source domain adaptation framework based on collaborative learning for semantic segmentation. Firstly, a simple image translation method is introduced to align t...
Data-driven based approaches, in spite of great success in many tasks, have poor generalization when applied to unseen image domains, and require expensive cost of annotation especially for dense pixel prediction tasks such as semantic segmentation. Recently, both unsupervised domain adaptation (UDA) from large amounts of synthetic data and semi-su...
Traditional neural architecture search (NAS) has a significant impact in computer vision by automatically designing network architectures for various tasks. In this paper, binarized neural architecture search (BNAS), with a search space of binarized convolutions, is introduced to produce extremely compressed models to reduce huge computational cost...
Deep model-based semantic segmentation has received ever increasing research focus in recent years. However, due to the complex model architectures, existing works are still unable to achieve high accuracy in real-time applications. In this paper, we propose a novel Sequential Prediction Network (termed SPNet) to seek a better trade-off between acc...
Natural language moment localization aims at localizing video clips according to a natural language description. The key to this challenging task lies in modeling the relationship between verbal descriptions and visual contents. Existing approaches often sample a number of clips from the video, and individually determine how each of them is related...
Person re-identification is a crucial task of identifying pedestrians of interest across multiple surveillance camera views. For person re-identification, a pedestrian is usually represented with features extracted from a rectangular image region that inevitably contains the scene background, which incurs ambiguity to distinguish different pedestri...
Citations
... Image Restoration (IR) is a fundamental vision task in the computer vision community, which aims to reconstruct a high-quality image from a degraded one. Recent advancements in deep learning have shown promising results in specific IR tasks such as denoising [8,16,78], dehazing [22,46,58], deraining [29,34,57], desnowing [12,13,48], motion deblurring [19,35,74], defocus deblurring [36, 61,81], low-light enhancement [6,25,69], and JPEG artifact removal/correction [33,86]. However, these models are limited to addressing only one specific type of degradation. ...
![[object Object]](https://i1.rgstatic.net/ii/profile.image/279619296088087-1443677843725_Q64/Songcen-Xu.jpg)
![[object Object]](https://i1.rgstatic.net/ii/profile.image/11431281213836286-1703210974025_Q64/Shuaicheng-Liu.jpg)
... In [30], the authors proposed a model that first extracts features from input images by multi-scale encoding modules and then produces an HDR image by progressively dilated U-shape blocks. [31] demonstrated that the ghosting problem is mainly in short-frequency signals, and therefore, they proposed a wavelet-based model to merge images in the frequency domain and avoid any ghosting problems. [32] implemented an algorithm that extracted dynamic areas of the images with the help of image segmentation and applied two neural networks separately on the static and dynamic scenes. ...
... The combination of the textual and visual inference could also be improved, either at the embedding level, following the idea of Xing et al. [2019], or at the score level, by learning a proper stacking or ensembling model. In addition, generic multi-class calibration methods could be used Guo et al. [2017], as well as methods adapted to the zero-shot LeVine et al. [2023] or the few-shot Wang et al. [2023] setups, which could help mitigate a potential domain shift. ...
![[object Object]](https://i1.rgstatic.net/ii/profile.image/272747213357071-1442039411273_Q64/Baochang-Zhang-2.jpg)
... Stable Diffusion (SD), an open-source text-to-image diffusion model trained on a large-scale online dataset [57], has emerged as a prominent choice for a diverse range of generative tasks and inverse problems. These tasks include but are not limited to image editing [1,2,21,30,69], inpainting [40,50,53], super-resolution [17,50,55], and image-to-image translation [10,42,76,77]. Despite the promising performance exhibited by SD, it encounters limitations when generating images at higher resolutions beyond its training resolution. ...
![[object Object]](https://i1.rgstatic.net/ii/profile.image/11431281157266997-1683758905513_Q64/Sheng-Xu-54.jpg)
... arXiv:2406.05658v1 [cs.CV] 9 Jun 2024 2 Related Work Prompting-Based Approaches: Most of the prompting-based approaches adopt a two-stage framework [37,39,14,15,32,34,35,11,18,19]: querying a group of prompts for an individual sample and using them to prompt the pre-trained models. For example, L2P [40] first selects a group of prompts from a prompt pool and then feeds them into the ViT. ...
... To circumvent this problem, some existing de-weathering methods (Valanarasu, Yasarla, and Patel 2022;Wang, Ma, and Liu 2023) initially construct paired data by simulating weather degradation and subsequently undergo super-vised learning. However, synthesized degraded images have limited realism in modeling complex and variable weather characteristics (e.g. ...
... To the best of our knowledge, we are the first to explore how test-time data can be leveraged in a continual learning setting to reduce forgetting. We consider the foundation model CLIP [40] for our experiments since it has been shown to encompass an extensive knowledge base and offer remarkable transferability [42,37]. It undergoes through supervised and unsupervised sessions, leveraging the unsupervised data to control forgetting. ...
... The advancement of deep generative models, including Generative Adversarial Networks (GANs) [3] and diffusion models [4], has facilitated the successful deployment of convenient and high-quality face reenactment techniques. A typical line of the state-of-the-art GANs or diffusion models based methods [5,6,7,8] relies on flow fields to complete the transferring of the pose and the expression. Initially, keypoints of the source image and the driving video are extracted in an unsupervised manner, followed by the estimation of motion flow fields based on these keypoints to capture pose and expression changes. ...
... As shown in Fig. 1, the feedforward path of the decoder is illustrated with black lines. The up-sampling operations in the decoder are implemented via PixelShuffle [31], followed by residual blocks for feature reconstruction. The output of the decoder after the feedforward process, dubbed I d , is obtained by progressive reconstruction from the output of DGAM. ...
... al. [2022] introduced spatial-varying operations, considering Signal-to-Noise Ratio (SNR) as a prior factor. However, when applying image enhancement algorithms to individual frames, the issue of flickering often arises. branch network that simultaneously estimates noise and illumination, suitable for videos with severe noise conditions. Liu et.al.Liu et al. [2023] and Liang et.al. Liang et al. [2023] used prior event information to learn enhancement mapping for brightening videos.Xu et.al. Xu et al. [2023a] designed a parametric 3D filter tailored for enhancing and sharpening low-light videos.Recently, Fu et.al. Fu et al. [2023a] introduced a video enhancement method called LAN, which iteratively ref ...
