Searcharxiv⌕ Search

arXiv subjects

Qi Zhao

Publications and source records attributed to Qi Zhao.

At least 163 records · Page 9Linked to original sources
arXiv2023

OV-VG: A Benchmark for Open-Vocabulary Visual Grounding

Open-vocabulary learning has emerged as a cutting-edge research area, particularly in light of the widespread adoption of vision-based foundational models. Its primary objective is to comprehend novel concepts that are not encompassed within a predefined vocabulary. One key facet of this endeavor is Visual Grounding, which entails locating a specific region within an image based on a corresponding language description. While current foundational models excel at various visual language tasks, there's a noticeable absence of models specifically tailored for open-vocabulary visual grounding. This research endeavor introduces novel and challenging OV tasks, namely Open-Vocabulary Visual Grounding and Open-Vocabulary Phrase Localization. The overarching aim is to establish connections between language descriptions and the localization of novel objects. To facilitate this, we have curated a comprehensive annotated benchmark, encompassing 7,272 OV-VG images and 1,000 OV-PL images. In our pursuit of addressing these challenges, we delved into various baseline methodologies rooted in existing open-vocabulary object detection, VG, and phrase localization frameworks. Surprisingly, we discovered that state-of-the-art methods often falter in diverse scenarios. Consequently, we developed a novel framework that integrates two critical components: Text-Image Query Selection and Language-Guided Feature Attention. These modules are designed to bolster the recognition of novel categories and enhance the alignment between visual and linguistic information. Extensive experiments demonstrate the efficacy of our proposed framework, which consistently attains SOTA performance across the OV-VG task. Additionally, ablation studies provide further evidence of the effectiveness of our innovative models. Codes and datasets will be made publicly available at https://github.com/cv516Buaa/OV-VG.

cs.CV↗
arXiv2023

What Do Deep Saliency Models Learn about Visual Attention?

In recent years, deep saliency models have made significant progress in predicting human visual attention. However, the mechanisms behind their success remain largely unexplained due to the opaque nature of deep neural networks. In this paper, we present a novel analytic framework that sheds light on the implicit features learned by saliency models and provides principled interpretation and quantification of their contributions to saliency prediction. Our approach decomposes these implicit features into interpretable bases that are explicitly aligned with semantic attributes and reformulates saliency prediction as a weighted combination of probability maps connecting the bases and saliency. By applying our framework, we conduct extensive analyses from various perspectives, including the positive and negative weights of semantics, the impact of training data and architectural designs, the progressive influences of fine-tuning, and common failure patterns of state-of-the-art deep saliency models. Additionally, we demonstrate the effectiveness of our framework by exploring visual attention characteristics in various application scenarios, such as the atypical attention of people with autism spectrum disorder, attention to emotion-eliciting stimuli, and attention evolution over time. Our code is publicly available at \url{https://github.com/szzexpoi/saliency_analysis}.

cs.CV↗
arXiv2023

Iterative Robust Visual Grounding with Masked Reference based Centerpoint Supervision

Visual Grounding (VG) aims at localizing target objects from an image based on given expressions and has made significant progress with the development of detection and vision transformer. However, existing VG methods tend to generate false-alarm objects when presented with inaccurate or irrelevant descriptions, which commonly occur in practical applications. Moreover, existing methods fail to capture fine-grained features, accurate localization, and sufficient context comprehension from the whole image and textual descriptions. To address both issues, we propose an Iterative Robust Visual Grounding (IR-VG) framework with Masked Reference based Centerpoint Supervision (MRCS). The framework introduces iterative multi-level vision-language fusion (IMVF) for better alignment. We use MRCS to ahieve more accurate localization with point-wised feature supervision. Then, to improve the robustness of VG, we also present a multi-stage false-alarm sensitive decoder (MFSD) to prevent the generation of false-alarm objects when presented with inaccurate expressions. The proposed framework is evaluated on five regular VG datasets and two newly constructed robust VG datasets. Extensive experiments demonstrate that IR-VG achieves new state-of-the-art (SOTA) results, with improvements of 25\% and 10\% compared to existing SOTA approaches on the two newly proposed robust VG datasets. Moreover, the proposed framework is also verified effective on five regular VG datasets. Codes and models will be publicly at https://github.com/cv516Buaa/IR-VG.

cs.CV↗
arXiv2023

Self-Training Guided Disentangled Adaptation for Cross-Domain Remote Sensing Image Semantic Segmentation

Deep convolutional neural networks (DCNNs) based remote sensing (RS) image semantic segmentation technology has achieved great success used in many real-world applications such as geographic element analysis. However, strong dependency on annotated data of specific scene makes it hard for DCNNs to fit different RS scenes. To solve this problem, recent works gradually focus on cross-domain RS image semantic segmentation task. In this task, different ground sampling distance, remote sensing sensor variation and different geographical landscapes are three main factors causing dramatic domain shift between source and target images. To decrease the negative influence of domain shift, we propose a self-training guided disentangled adaptation network (ST-DASegNet). We first propose source student backbone and target student backbone to respectively extract the source-style and target-style feature for both source and target images. Towards the intermediate output feature maps of each backbone, we adopt adversarial learning for alignment. Then, we propose a domain disentangled module to extract the universal feature and purify the distinct feature of source-style and target-style features. Finally, these two features are fused and served as input of source student decoder and target student decoder to generate final predictions. Based on our proposed domain disentangled module, we further propose exponential moving average (EMA) based cross-domain separated self-training mechanism to ease the instability and disadvantageous effect during adversarial optimization. Extensive experiments and analysis on benchmark RS datasets show that ST-DASegNet outperforms previous methods on cross-domain RS image semantic segmentation task and achieves state-of-the-art (SOTA) results. Our code is available at https://github.com/cv516Buaa/ST-DASegNet.

cs.CV↗
arXiv2023

Change Detection Methods for Remote Sensing in the Last Decade: A Comprehensive Review

Change detection is an essential and widely utilized task in remote sensing that aims to detect and analyze changes occurring in the same geographical area over time, which has broad applications in urban development, agricultural surveys, and land cover monitoring. Detecting changes in remote sensing images is a complex challenge due to various factors, including variations in image quality, noise, registration errors, illumination changes, complex landscapes, and spatial heterogeneity. In recent years, deep learning has emerged as a powerful tool for feature extraction and addressing these challenges. Its versatility has resulted in its widespread adoption for numerous image-processing tasks. This paper presents a comprehensive survey of significant advancements in change detection for remote sensing images over the past decade. We first introduce some preliminary knowledge for the change detection task, such as problem definition, datasets, evaluation metrics, and transformer basics, as well as provide a detailed taxonomy of existing algorithms from three different perspectives: algorithm granularity, supervision modes, and learning frameworks in the methodology section. This survey enables readers to gain systematic knowledge of change detection tasks from various angles. We then summarize the state-of-the-art performance on several dominant change detection datasets, providing insights into the strengths and limitations of existing algorithms. Based on our survey, some future research directions for change detection in remote sensing are well identified. This survey paper will shed some light on the community and inspire further research efforts in the change detection task.

cs.CV↗
arXiv2023

DNeRV: Modeling Inherent Dynamics via Difference Neural Representation for Videos

Existing implicit neural representation (INR) methods do not fully exploit spatiotemporal redundancies in videos. Index-based INRs ignore the content-specific spatial features and hybrid INRs ignore the contextual dependency on adjacent frames, leading to poor modeling capability for scenes with large motion or dynamics. We analyze this limitation from the perspective of function fitting and reveal the importance of frame difference. To use explicit motion information, we propose Difference Neural Representation for Videos (DNeRV), which consists of two streams for content and frame difference. We also introduce a collaborative content unit for effective feature fusion. We test DNeRV for video compression, inpainting, and interpolation. DNeRV achieves competitive results against the state-of-the-art neural compression approaches and outperforms existing implicit methods on downstream inpainting and interpolation for $960 \times 1920$ videos.

cs.CV↗
arXiv2023

A Self-Distillation Embedded Supervised Affinity Attention Model for Few-Shot Segmentation

Few-shot segmentation focuses on the generalization of models to segment unseen object with limited annotated samples. However, existing approaches still face two main challenges. First, huge feature distinction between support and query images causes knowledge transferring barrier, which harms the segmentation performance. Second, limited support prototypes cannot adequately represent features of support objects, hard to guide high-quality query segmentation. To deal with the above two issues, we propose self-distillation embedded supervised affinity attention model to improve the performance of few-shot segmentation task. Specifically, the self-distillation guided prototype module uses self-distillation to align the features of support and query. The supervised affinity attention module generates high-quality query attention map to provide sufficient object information. Extensive experiments prove that our model significantly improves the performance compared to existing methods. Comprehensive ablation experiments and visualization studies also show the significant effect of our method on few-shot segmentation task. On COCO-20i dataset, we achieve new state-of-the-art results. Training code and pretrained models are available at https://github.com/cv516Buaa/SD-AANet.

cs.CV↗
arXiv2023

Divide and Conquer: Answering Questions with Object Factorization and Compositional Reasoning

Humans have the innate capability to answer diverse questions, which is rooted in the natural ability to correlate different concepts based on their semantic relationships and decompose difficult problems into sub-tasks. On the contrary, existing visual reasoning methods assume training samples that capture every possible object and reasoning problem, and rely on black-boxed models that commonly exploit statistical priors. They have yet to develop the capability to address novel objects or spurious biases in real-world scenarios, and also fall short of interpreting the rationales behind their decisions. Inspired by humans' reasoning of the visual world, we tackle the aforementioned challenges from a compositional perspective, and propose an integral framework consisting of a principled object factorization method and a novel neural module network. Our factorization method decomposes objects based on their key characteristics, and automatically derives prototypes that represent a wide range of objects. With these prototypes encoding important semantics, the proposed network then correlates objects by measuring their similarity on a common semantic space and makes decisions with a compositional reasoning process. It is capable of answering questions with diverse objects regardless of their availability during training, and overcoming the issues of biased question-answer distributions. In addition to the enhanced generalizability, our framework also provides an interpretable interface for understanding the decision-making process of models. Our code is available at https://github.com/szzexpoi/POEM.

cs.CV↗
arXiv2023

Unextendible product bases from orthogonality graphs

Unextendible product bases (UPBs) play a key role in the study of quantum entanglement and nonlocality. A famous open question is whether there exist genuinely unextendible product bases (GUPBs), namely multipartite product bases that are unextendible with respect to every possible bipartition. Here we shed light on this question by providing a characterization of UPBs and GUPBs in terms of orthogonality graphs. Building on this connection, we develop a method for constructing UPBs in low dimensions, and we derive a lower bound on the size of any GUPB, significantly improving over the state of the art. Moreover, we show that every minimal GUPB saturating our bound must be associated to regular graphs. Finally, we discuss a possible path towards the construction of a minimal GUPB in a tripartite system of minimal local dimension.

quant-ph↗
arXiv2023

Transformation and reconstruction towards two-dimensional atomic laminates

Two-dimensional (2D) nanomaterials derived from non-van der Waals solids are promising due to their fantastic physical and chemical properties, but it remains challenging to obtain 2D atomic laminates with high stability owing to the strong intrinsic covalent/metallic bonds and highly exposed active surface. Here, we report a versatile and scalable protocol to produce 2D atomic laminates, based on an unexpected topological transformation of MAX phases under hydrogen chloride gas, and/or subsequent reconstruction under some programmed gases/vapors. In contrast to the known approaches with liquid or molten medium, our method involves in a gas-phase reaction with fast thermodynamics for A layers and positive Gibbs free energies for MX slabs. Remarkably, through subsequent reconstruction in some active gases/vapors (O2, H2S, P, CH4, Al and Sn metal vapors), a big family of 2D atomic laminates with elusive configurations as well as high chemical/thermal stabilities and tunable electrical properties (from metallic to semiconductor-like behaviors) are achieved. Moreover, the resultant 2D atomic laminates can be facilely scaled up to 10 kilograms. We believe that the 2D atomic laminates would have broad applications in catalysis, energy storage, electromagnetic shielding interface and microwave absorption.

cond-mat.mtrl-sci↗
arXiv2023

Metal-bonded Atomic Layers of Transition Metal Carbides (MXenes)

Although two-dimensional transition metal carbides and nitrides (MXenes) have fantastic physical and chemical properties as well as wide applications, it remains challenging to produce stable MXenes due to their rapid structural degradation. Here, unique metal-bonded atomic layers of transition metal carbides with high stabilities are produced via a simple topological reaction between chlorine-terminated MXenes and selected metals, where the metals enable to not only remove Cl terminations, but also efficiently bond with adjacent atomic MXene slabs, driven by the symmetry of MAX phases. The films constructed from Al-bonded Ti$_3$C$_2$Cl$_x$ atomic layers show high oxidation resistance up to 400 degrees centigrade and low sheet resistance of 9.3 ohm per square. Coupled to the multi-layer structure, the Al-bonded Ti$_3$C$_2$Cl$_x$ film displays a significantly improved EMI shielding capability with a total shielding effectiveness value of 39 dB at a low thickness of 3.1 micron, outperforming pure Ti$_3$C$_2$Cl$_x$ film.

cond-mat.mtrl-sci↗
arXiv2023

Numerical investigation of progressive damage and associated seismicity on a laboratory fault

Understanding rock shear failure behavior is crucial to gain insights into slip-related geohazards such as rock avalanches, landslides, and earthquakes. However, descriptions of the progressive damage on the shear surface are still incomplete or ambiguous. In this study, we use the hybrid finite-discrete element method (FDEM) to simulate a shear experiment and obtain a detailed comprehension of shear induced progressive damage and the associated seismic activity. We built a laboratory fault model from high resolution surface scans and micro-CT imaging. Our results show that under quasi-static shear loading, the fault surface experiences local dynamic seismic activities. We found that the seismic activity is related to the stress concentration on interlocking asperities. This interlocking behavior (i) causes stress concentration at the region of contact that could reach the compressive strength, and (ii) produces tensile stress up to the tensile strength in the region adjacent to the contact area. Thus, different failure mechanisms and damage patterns including crushing and sub-vertical fracturing are observed on the rough surface. Asperity failure creates rapid local slips resulting in significant stress perturbations that alter the overall stress condition and may trigger the slip of adjacent critically stressed asperities. We found that the spatial distribution of the damaged asperities and the seismic activity is highly heterogeneous; regions with intense asperity interactions formed gouge material, while others exhibit minimal to no damage. These results emphasize the important role of surface roughness in controlling the overall shear behavior and the local dynamic seismic activities on faults.

physics.geo-ph↗
arXiv2022

Towards efficient and generic entanglement detection by machine learning

Detection of entanglement is an indispensable step to practical quantum computation and communication. Compared with the conventional entanglement witness method based on fidelity, we propose a flexible, machine learning assisted entanglement detection protocol that is robust to different types of noises and sample efficient. In this protocol, an entanglement classifier for a generic entangled state is obtained by training a classical machine learning model with a synthetic dataset. The dataset contains classical features of two types of states and their labels (either entangled or separable). The classical features of a state, which are expectation values of a set of k-local Pauli observables, are estimated sample-efficiently by the classical shadow method. In the numerical simulation, our classifier can detect the entanglement of 4-qubit GHZ states with coherent noise and W states mixed with large white noise, with high accuracy.

quant-ph↗
arXiv2022

A scheme to create and verify scalable entanglement in optical lattice

To achieve scalable quantum information processing, great efforts have been devoted to the creation of large-scale entangled states in various physical systems. Ultracold atom in optical lattice is considered as one of the promising platforms due to its feasible initialization and parallel manipulation. In this work, we propose an efficient scheme to generate and characterize global entanglement in the optical lattice. With only two-layer quantum circuits, the generation utilizes two-qubit entangling gates based on the superexchange interaction in double wells. The parallelism of these operations enables the generation to be fast and scalable. To verify the entanglement of this non-stabilizer state, we mainly design three complementary detection protocols which are less resource-consuming compared to the full tomography. In particular, one just needs two homogenous local measurement settings to identify the entanglement property. Our entanglement generation and verification protocols provide the foundation for the further quantum information processing in optical lattice.

quant-ph↗
arXiv2022

Improved Pump Setpoint Selection Using a Calibrated Hydraulic Model of a High-Pressure Irrigation System

This paper presents a case study of the operational management of the Robinvale high-pressure piped irrigation water delivery system (RVHPS) in Australia. Based on datasets available, improved pump setpoint selection using a calibrated hydraulic model is investigated. The first step was to implement pre-processing of measured flow and pressure data to identify errors in the data and possible faulty sensors. An EPANET hydraulic simulation model was updated with calibrated pipe roughness height values by using the processed pressure and flow data. Then, new pump setpoints were selected using the calibrated model given the actual measured demands such that the pressures in the network were minimized subject to required customer service standards. Based on a two-day simulation, it was estimated that 4.7% savings in pumping energy cost as well as 4.7% reduction in greenhouse gas emissions can be achieved by applying the new pump setpoints.

eess.SY↗
arXiv2022

Look in Different Views: Multi-Scheme Regression Guided Cell Instance Segmentation

Cell instance segmentation is a new and challenging task aiming at joint detection and segmentation of every cell in an image. Recently, many instance segmentation methods have applied in this task. Despite their great success, there still exists two main weaknesses caused by uncertainty of localizing cell center points. First, densely packed cells can easily be recognized into one cell. Second, elongated cell can easily be recognized into two cells. To overcome these two weaknesses, we propose a novel cell instance segmentation network based on multi-scheme regression guidance. With multi-scheme regression guidance, the network has the ability to look each cell in different views. Specifically, we first propose a gaussian guidance attention mechanism to use gaussian labels for guiding the network's attention. We then propose a point-regression module for assisting the regression of cell center. Finally, we utilize the output of the above two modules to further guide the instance segmentation. With multi-scheme regression guidance, we can take full advantage of the characteristics of different regions, especially the central region of the cell. We conduct extensive experiments on benchmark datasets, DSB2018, CA2.5 and SCIS. The encouraging results show that our network achieves SOTA (state-of-the-art) performance. On the DSB2018 and CA2.5, our network surpasses previous methods by 1.2% (AP50). Particularly on SCIS dataset, our network performs stronger by large margin (3.0% higher AP50). Visualization and analysis further prove that our proposed method is interpretable.

cs.CV↗
arXiv2022

Unextendible and uncompletable product bases in every bipartition

Unextendible product basis is an important object in quantum information theory and features a broad spectrum of applications, ranging bound entangled states, quantum nonlocality without entanglement, and Bell inequalities with no quantum violation. A generalized concept called uncompletable product basis also attracts much attention. In this paper, we find some unextendible product bases that are uncompletable product bases in every bipartition, which answers a 19 year-old open question proposed by DiVincenzo et al. [Commun. Math. Phys. 238, 379 (2003)]. As a consequence, we connect such unextendible product bases to local hiding of information and give a sufficient condition for the existence of an unextendible product basis, that is still an unextendible product basis in every bipartition. Our results advance the understanding of the geometry of unextendible product bases.

quant-ph↗
arXiv2022

Constrained Heterogeneous Two-facility Location Games with Max-variant Cost

In this paper, we propose a constrained heterogeneous facility location model where a set of alternative locations are feasible for building facilities and the number of facilities built at each location is limited. Supposing that a set of agents on the real line can strategically report their locations and each agent's cost is her distance to the further facility that she is interested in, we study deterministic mechanism design without money for constrained heterogeneous two-facility location games. Depending on whether agents have optional preference, the problem is considered in two settings: the compulsory setting and the optional setting. In the compulsory setting where each agent is served by the two heterogeneous facilities, we provide a 3-approximate deterministic group strategyproof mechanism for the sum/maximum cost objective respectively, which is also the best deterministic strategyproof mechanism under the corresponding social objective. In the optional setting where each agent can be interested in one of the two facilities or both, we propose a deterministic group strategyproof mechanism with approximation ratio of at most $2n+1$ for the sum cost objective and a deterministic group strategyproof mechanism with approximation ratio of at most 9 for the maximum cost objective.

cs.GT↗