Careful robot manipulation in every-day cluttered environments requires an accurate understanding of the 3D scene, in order to grasp and place objects stably and reliably and to avoid mistakenly colliding with other objects. In general, we must construct such a 3D interpretation of a complex scene based on limited input, such as a single RGB-D image. We describe SceneComplete, a system for constructing a complete, segmented, 3D model of a scene from a single view. It provides a novel pipeline for composing general-purpose pretrained perception modules (vision-language, segmentation, image-inpainting, image-to-3D, and pose-estimation) to obtain high-accuracy results. We demonstrate its accuracy and effectiveness with respect to ground-truth models in a large benchmark dataset and show that its accurate whole-object reconstruction enables robust grasp proposal generation, including for a dexterous hand.
相關內容
Significant achievements in personalization of diffusion models have been witnessed. Conventional tuning-free methods mostly encode multiple reference images by averaging their image embeddings as the injection condition, but such an image-independent operation cannot perform interaction among images to capture consistent visual elements within multiple references. Although the tuning-based Low-Rank Adaptation (LoRA) can effectively extract consistent elements within multiple images through the training process, it necessitates specific finetuning for each distinct image group. This paper introduces EasyRef, a novel plug-and-play adaptation method that enables diffusion models to be conditioned on multiple reference images and the text prompt. To effectively exploit consistent visual elements within multiple images, we leverage the multi-image comprehension and instruction-following capabilities of the multimodal large language model (MLLM), prompting it to capture consistent visual elements based on the instruction. Besides, injecting the MLLM's representations into the diffusion process through adapters can easily generalize to unseen domains, mining the consistent visual elements within unseen data. To mitigate computational costs and enhance fine-grained detail preservation, we introduce an efficient reference aggregation strategy and a progressive training scheme. Finally, we introduce MRBench, a new multi-reference image generation benchmark. Experimental results demonstrate EasyRef surpasses both tuning-free methods like IP-Adapter and tuning-based methods like LoRA, achieving superior aesthetic quality and robust zero-shot generalization across diverse domains.
To address the challenges of robust data transmission over complex time-varying channels, this paper introduces channel learning and enhanced adaptive reconstruction (CLEAR) strategy for semantic communications. CLEAR integrates deep joint source-channel coding (DeepJSCC) with an adaptive diffusion denoising model (ADDM) to form a unique framework. It leverages a trainable encoder-decoder architecture to encode data into complex semantic codes, which are then transmitted and reconstructed while minimizing distortion, ensuring high semantic fidelity. By addressing multipath effects, frequency-selective fading, phase noise, and Doppler shifts, CLEAR achieves high semantic fidelity and reliable transmission across diverse signal-to-noise ratios (SNRs) and channel conditions. Extensive experiments demonstrate that CLEAR achieves a 2.3 dB gain on peak signal-to-noise ratio (PSNR) over the existing state-of-the-art method, DeepJSCC-V. Furthermore, the results verify that CLEAR is robust against varying channel conditions, particularly in scenarios characterized by high Doppler shifts and strong phase noise.
Recent advances in deep learning research have shown remarkable achievements across many tasks in computer vision (CV) and natural language processing (NLP). At the intersection of CV and NLP is the problem of image captioning, where the related models' robustness against adversarial attacks has not been well studied. This paper presents a novel adversarial attack strategy, AICAttack (Attention-based Image Captioning Attack), designed to attack image captioning models through subtle perturbations on images. Operating within a black-box attack scenario, our algorithm requires no access to the target model's architecture, parameters, or gradient information. We introduce an attention-based candidate selection mechanism that identifies the optimal pixels to attack, followed by a customised differential evolution method to optimise the perturbations of pixels' RGB values. We demonstrate AICAttack's effectiveness through extensive experiments on benchmark datasets against multiple victim models. The experimental results demonstrate that our method outperforms current leading-edge techniques by achieving consistently higher attack success rates.
Soft robots have the potential to interact with sensitive environments and perform complex tasks effectively. However, motion plans and trajectories for soft manipulators are challenging to calculate due to their deformable nature and nonlinear dynamics. This article introduces a fast real-time trajectory generation approach for soft robot manipulators, which creates dynamically-feasible motions for arbitrary kinematically-feasible paths of the robot's end effector. Our insight is that piecewise constant curvature (PCC) dynamics models of soft robots can be differentially flat, therefore control inputs can be calculated algebraically rather than through a nonlinear differential equation. We prove this flatness under certain conditions, with the curvatures of the robot as the flat outputs. Our two-step trajectory generation approach uses an inverse kinematics procedure to calculate a motion plan of robot curvatures per end-effector position, then, our flatness diffeomorphism generates corresponding control inputs that respect velocity. We validate our approach through simulations of our representative soft robot manipulator along three different trajectories, demonstrating a margin of 23x faster than real-time at a frequency of 100 Hz. This approach could allow fast verifiable replanning of soft robots' motions in safety-critical physical environments, crucial for deployment in the real world.
Abstract representations of 3D scenes play a crucial role in computer vision, enabling a wide range of applications such as mapping, localization, surface reconstruction, and even advanced tasks like SLAM and rendering. Among these representations, line segments are widely used because of their ability to succinctly capture the structural features of a scene. However, existing 3D reconstruction methods often face significant challenges. Methods relying on 2D projections suffer from instability caused by errors in multi-view matching and occlusions, while direct 3D approaches are hampered by noise and sparsity in 3D point cloud data. This paper introduces LineGS, a novel method that combines geometry-guided 3D line reconstruction with a 3D Gaussian splatting model to address these challenges and improve representation ability. The method leverages the high-density Gaussian point distributions along the edge of the scene to refine and optimize initial line segments generated from traditional geometric approaches. By aligning these segments with the underlying geometric features of the scene, LineGS achieves a more precise and reliable representation of 3D structures. The results show significant improvements in both geometric accuracy and model compactness compared to baseline methods.
Deep Learning Training (DLT) is a growing workload in shared GPU/CPU clusters due to its high computational cost and increasing number of jobs. This contributes to significant energy consumption in GPU clusters, further exacerbated by GPU under-utilization, as shown in production cluster logs. Addressing this challenge requires workload scheduling and resource allocation policies for efficient GPU sharing to improve resource and energy efficiency while maintaining performance. However, previous works primarily optimize for performance, often overlooking or even sacrificing energy efficiency. In this paper, we present EaCO, the first energy-aware scheduling algorithm designed specifically for DLT workloads in GPU clusters. EaCO leverages hardware-supported context switching to enable GPU sharing across multiple DLT jobs, improving resource and energy utilization. GPU sharing can increase Job Completion Time (JCT) and may lead to contention if not employed carefully. To address this, EaCO integrates experiment and historical-based predictions as well as early-stage observations, ensuring performance expectations are met while optimizing energy efficiency. We begin by experimentally exploring the dynamics of co-locating DLTs, investigating its impact on energy and resource utilization. Our results show that co-location improves energy efficiency by up to 44% for individual jobs, and increases average GPU utilization to as high as 97%. Additionally, evaluations on large-scale clusters using production traces demonstrate that EaCO reduces total energy by up to 39% compared to existing algorithms, which comes with a minimal increase in job runtime-less than 3.2% in our simulations.
This paper aims to manipulate multi-entity 3D motions in video generation. Previous methods on controllable video generation primarily leverage 2D control signals to manipulate object motions and have achieved remarkable synthesis results. However, 2D control signals are inherently limited in expressing the 3D nature of object motions. To overcome this problem, we introduce 3DTrajMaster, a robust controller that regulates multi-entity dynamics in 3D space, given user-desired 6DoF pose (location and rotation) sequences of entities. At the core of our approach is a plug-and-play 3D-motion grounded object injector that fuses multiple input entities with their respective 3D trajectories through a gated self-attention mechanism. In addition, we exploit an injector architecture to preserve the video diffusion prior, which is crucial for generalization ability. To mitigate video quality degradation, we introduce a domain adaptor during training and employ an annealed sampling strategy during inference. To address the lack of suitable training data, we construct a 360-Motion Dataset, which first correlates collected 3D human and animal assets with GPT-generated trajectory and then captures their motion with 12 evenly-surround cameras on diverse 3D UE platforms. Extensive experiments show that 3DTrajMaster sets a new state-of-the-art in both accuracy and generalization for controlling multi-entity 3D motions. Project page: //fuxiao0719.github.io/projects/3dtrajmaster
With the development of science and technology, mobile robots are playing a significant important role in the new round of world revolution. Further, mobile robots might assist or replace human beings in a great number of areas. To increase the degree of automation for mobile robots, advanced motion planners need to be integrated into them to cope with various environments. Complex maze environments are common in the potential application scenarios of different mobile robots. This article proposes a novel motion planner named the rapidly exploring random tree based Gaussian process motion planner 2, which aims to tackle the motion planning problem for mobile robots in complex maze environments. To be more specific, the proposed motion planner successfully combines the advantages of a trajectory optimisation motion planning algorithm named the Gaussian process motion planner 2 and a sampling-based motion planning algorithm named the rapidly exploring random tree. To validate the performance and practicability of the proposed motion planner, we have tested it in several simulations in the Matrix laboratory and applied it on a marine mobile robot in a virtual scenario in the Robotic operating system.
Objective:To develop a no-reference image quality assessment method using automated distortion recognition to boost MRI-guided radiotherapy precision.Methods:We analyzed 106,000 MR images from 10 patients with liver metastasis,captured with the Elekta Unity MR-LINAC.Our No-Reference Quality Assessment Model includes:1)image preprocessing to enhance visibility of key diagnostic features;2)feature extraction and directional analysis using MSCN coefficients across four directions to capture textural attributes and gradients,vital for identifying image features and potential distortions;3)integrative Quality Index(QI)calculation,which integrates features via AGGD parameter estimation and K-means clustering.The QI,based on a weighted MAD computation of directional scores,provides a comprehensive image quality measure,robust against outliers.LOO-CV assessed model generalizability and performance.Tumor tracking algorithm performance was compared with and without preprocessing to verify tracking accuracy enhancements.Results:Preprocessing significantly improved image quality,with the QI showing substantial positive changes and surpassing other metrics.After normalization,the QI's average value was 79.6 times higher than CNR,indicating improved image definition and contrast.It also showed higher sensitivity in detail recognition with average values 6.5 times and 1.7 times higher than Tenengrad gradient and entropy.The tumor tracking algorithm confirmed significant tracking accuracy improvements with preprocessed images,validating preprocessing effectiveness.Conclusions:This study introduces a novel no-reference image quality evaluation method based on automated distortion recognition,offering a new quality control tool for MRIgRT tumor tracking.It enhances clinical application accuracy and facilitates medical image quality assessment standardization, with significant clinical and research value.
Multimodal multihop question answering is a complex task that requires reasoning over multiple sources of information, such as images and text, to answer questions. While there has been significant progress in visual question answering, the multihop setting remains unexplored due to the lack of high-quality datasets. Current methods focus on single-hop question answering or a single modality, which makes them unsuitable for real-world scenarios such as analyzing multimodal educational materials, summarizing lengthy academic articles, or interpreting scientific studies that combine charts, images, and text. To address this gap, we propose a novel methodology, introducing the first framework for creating a high-quality dataset that enables training models for multimodal multihop question answering. Our approach consists of a 5-stage pipeline that involves acquiring relevant multimodal documents from Wikipedia, synthetically generating high-level questions and answers, and validating them through rigorous criteria to ensure quality data. We evaluate our methodology by training models on our synthesized dataset and testing on two benchmarks, our results demonstrate that, with an equal sample size, models trained on our synthesized data outperform those trained on human-collected data by 1.9 in exact match (EM) on average. We believe our data synthesis method will serve as a strong foundation for training and evaluating multimodal multihop question answering models.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191