2023-11-27 Research Report

【1】Image Super-Resolution with Text Prompt Diffusion [arxiv code]

标题：结合文本实例扩散模型的图像超分

【2】RSB-Pose: Robust Short-Baseline Binocular 3D Human Pose Estimation with Occlusion Handling [IEEE TIP2024]

标题：带遮挡处理的鲁棒短基线双目三维人体姿态估计

【3】ECRF: Entropy-Constrained Neural Radiance Fields Compression with Frequency Domain Optimization [arxiv]

标题：熵约束神经辐射场压缩与频域优化

【4】HACD: Hand-Aware Conditional Diffusion for Monocular Hand-Held Object Reconstruction [arxiv]

标题：用于单目手持物体重构的手感条件扩散技术

【5】Language-guided Few-shot Semantic Segmentation [ICASSP2024]

标题：语言引导的少样本语义分割

【6】 Dynamic Compositional Graph Convolutional Network for Efficient Composite Human Motion Prediction [arxiv]

标题：用于高效复合人体运动预测的动态合成图卷积网络

【7】TRIDENT: The Nonlinear Trilogy for Implicit Neural Representations [arxiv]

标题：隐含神经表征的非线性三部曲

【8】Joint Diffusion: Mutual Consistency-Driven Diffusion Model for PET-MRI Co-Reconstruction [arxiv]

标题：联合扩散： 用于 PET-MRI 协同重建的相互一致性驱动的扩散模型

【9】ConvNet vs Transformer, Supervised vs CLIP: Beyond ImageNet Accuracy [arxiv]

标题：卷积 vs Transformer，监督 vs CLIP：超越ImageNet的准确性

【10】Domain Aligned CLIP for Few-shot Classification [WACV2024]

标题：少样本分类的CLIP域对齐

【11】Imagine the Unseen World: A Benchmark for Systematic Generalization in Visual World Models[NIPS2023]

标题：想象看不见的世界： 视觉世界模型中的系统泛化基准

【12】Frequency Domain-based Dataset Distillation[NIPS2023]

标题：频率域数据集蒸馏

【13】Topology of Surface Electromyogram Signals: Hand Gesture Decoding on Riemannian Manifolds [arxiv]

标题：表面肌电信号拓扑： 黎曼曲面上的手势解码

【14】CLIP Guided Image-perceptive Prompt Learning for Image Enhancement[arxiv]

标题：CLIP 引导图像感知提示学习以增强图像效果

【15】OW-SLR: Overlapping Windows on Semi-Local Region for Image Super-Resolution[arxiv]

标题：用于图像超分辨率的半局部区域重叠窗口

【16】Vision Transformer with Super Token Sampling [arxiv]

标题：结合超Token采样的视觉Transformer

【17】Versatile Medical Image Segmentation Learned from Multi-Source Datasets via Model Self-Disambiguation[arxiv]

标题：通过模型自消歧从多源数据集学习的多功能医学图像分割技术

【18】Segment Anything Model with Uncertainty Rectification for Auto-Prompting Medical Image Segmentation[arxiv]

标题：用于自动提示医学图像分割的带有不确定性校正功能的任意分割模型

【19】Breaking Temporal Consistency: Generating Video Universal Adversarial Perturbations Using Image Models[ICCV2023]

标题：打破时空一致性： 利用图像模型生成视频通用对抗扰动

【20】Clarity ChatGPT: An Interactive and Adaptive Processing System for Image Restoration and Enhancement[arxiv]

标题：Clarity ChatGPT： 用于图像修复和增强的交互式自适应处理系统

【21】Cut-and-Paste: Subject-Driven Video Editing with Attention Control[arxiv]

标题：剪切粘贴： 主题驱动的视频编辑与注意力控制

【22】Deep Equilibrium Diffusion Restoration with Parallel Sampling[arxiv]

标题：利用并行采样进行深度平衡扩散复原

【23】MoVideo: Motion-Aware Video Generation with Diffusion Models[arxiv project]

标题：MoVideo： 利用扩散模型生成运动感知视频

【24】Pair-wise Layer Attention with Spatial Masking for Video Prediction [arxiv]

标题：利用空间掩码的成对层级注意力进行视频预测

【25】MaskFlow: Object-Aware Motion Estimation[arxiv]

标题：MaskFlow：物体感知运动估计

【26】Learning Part Motion of Articulated Objects Using Spatially Continuous Neural Implicit Representations[BMVC2023]

标题：利用空间连续神经隐含表征学习关节物体的部分运动

【27】CASR: Refining Action Segmentation via Magrinalizing Frame-levle Causal Relationships[arxiv]

标题：通过边缘化帧级别的因果关系实现动作分割的精细化

【28】Disentangling Structure and Appearance in ViT Feature Space[TOG2023]

标题：在 ViT 特征空间中分离结构与外观

【29】Efficient Model Agnostic Approach for Implicit Neural Representation Based Arbitrary-Scale Image Super-Resolution[arxiv]

标题：基于隐式神经表征的任意尺度图像超分辨率的高效模型无关方法

【30】Swift Parameter-free Attention Network for Efficient Super-Resolution[arxiv]

标题：用于高效超分辨率的快速无参数注意力网络