We introduce Housekeep, a benchmark to evaluate commonsense reasoning in the home for embodied AI. In Housekeep, an embodied agent must tidy a house by rearranging misplaced objects without explicit instructions specifying which objects need to be rearranged. Instead, the agent must learn from and is evaluated against human preferences of which objects belong where in a tidy house. Specifically, we collect a dataset of where humans typically place objects in tidy and untidy houses constituting 1799 objects, 268 object categories, 585 placements, and 105 rooms. Next, we propose a modular baseline approach for Housekeep that integrates planning, exploration, and navigation. It leverages a fine-tuned large language model (LLM) trained on an internet text corpus for effective planning. We show that our baseline generalizes to rearranging unseen objects in unknown environments.
论文检索
输入标题、作者或关键词，从 1,645 篇学术成果中精准定位
We propose a domain adaptation method, MoDA, which adapts a pretrained embodied agent to a new, noisy environment without ground-truth supervision. Map-based memory provides important contextual information for visual navigation, and exhibits unique spatial structure mainly composed of flat walls and rectangular obstacles. Our adaptation approach encourages the inherent regularities on the estimated maps to guide the agent to overcome the prevalent domain discrepancy in a novel environment. Specifically, we propose an efficient learning curriculum to handle the visual and dynamics corruptions in an online manner, self-supervised with pseudo clean maps generated by style transfer networks. Because the map-based representation provides spatial knowledge for the agent’s policy, our formulation can deploy the pretrained policy networks from simulators in a new setting. We evaluate MoDA in various practical scenarios and show that our proposed method quickly enhances the agent’s performance in downstream tasks including localization, mapping, exploration, and point-goal navigation.
Object manipulation is a critical skill required for Embodied AI agents interacting with the world around them. Training agents to manipulate objects, poses many challenges. These include occlusion of the target object by the agent’s arm, noisy object detection and localization, and the target frequently going out of view as the agent moves around in the scene. We propose Manipulation via Visual Object Location Estimation (m-VOLE), an approach that explores the environment in search for target objects, computes their 3D coordinates once they are located, and then continues to estimate their 3D locations even when the objects are not visible, thus robustly aiding the task of manipulating these objects throughout the episode. Our evaluations show a massive 3x improvement in success rate over a model that has access to the same sensory suite but is trained without the object location estimator, and our analysis shows that our agent is robust to noise in depth perception and agent localization. Importantly, our proposed approach relaxes several assumptions about idealized localization and perception that are commonly employed by recent works in navigation and manipulation -- an important step towards training agents for object manipulation in the real world. Our code and data are available at https://prior.allenai.org/projects/m-vole}{prior.allenai.org/projects/m-vole.
Object navigation is a task to let the agent navigate to a target object. Prevailing works attempt to expand navigation ability in new environments and achieve reasonable performance on the seen object categories that have been observed in training environments. However, this setting is somewhat limited in real world scenario, where navigating to unseen object categories is generally unavoidable. In this paper, we focus on the problem of navigating to unseen objects in new environments only based on limited training knowledge. Same as the common ObjectNav tasks, our agent still gets the egocentric observation and target object category as the input and does not require any extra inputs. Our solution is to let the agent “imagine"" the unseen object by synthesizing features of the target object. We propose a generative meta-adversarial network (GMAN), which is mainly composed of a feature generator and an environmental meta discriminator, aiming to generate features for unseen objects and new environments in two steps. The former generates the initial features of the unseen objects based on the semantic embedding of the object category. The latter enables the generator to further learn the background characteristics of the new environment, progressively adapting the generated features to approximate the real features of the target object. The adapted features serve as a more specific representation of the target to guide the agent. Moreover, to fast update the generator with a few observations, the entire adversarial framework is learned in the gradient-based meta-learning manner. The experimental results on AI2THOR and RoboTHOR simulators demonstrate the effectiveness of the proposed method in navigating to unseen object categories.
LiDAR is widely used to capture accurate 3D outdoor scene structures. However, LiDAR produces many undesirable noise points in snowy weather, which hamper analyzing meaningful 3D scene structures. Semantic segmentation with snow labels would be a straightforward solution for removing them, but it requires laborious point-wise annotation. To address this problem, we propose a novel self-supervised learning framework for snow points removal in LiDAR point clouds. Our method exploits the structural characteristic of the noise points: low spatial correlation with their neighbors. Our method consists of two deep neural networks: Point Reconstruction Network (PR-Net) reconstructs each point from its neighbors; Reconstruction Difficulty Network (RD-Net) predicts point-wise difficulty of the reconstruction by PR-Net, which we call reconstruction difficulty. With simple post-processing, our method effectively detects snow points without any label. Our method achieves the state-of-the-art performance among label-free approaches and is comparable to the fully-supervised method. Moreover, we demonstrate that our method can be exploited as a pretext task to improve label-efficiency of supervised training of de-snowing.
Recent advances in high-fidelity simulators [22,82,44] have enabled closed-loop training of autonomous driving agents, potentially solving the distribution shift in training v.s. deployment and allowing training to be scaled both safely and cheaply. However, there is a lack of understanding of how to build effective training benchmarks for closed-loop training. In this work, we present the first empirical study which analyzes the effects of different training benchmark designs on the success of learning agents, such as how to design traffic scenarios and scale training environments. Furthermore, we show that many popular RL algorithms cannot achieve satisfactory performance in the context of autonomous driving, as they lack long-term planning and take an extremely long time to train. To address these issues, we propose trajectory value learning (TRAVL), an RL-based driving agent approach that performs planning with multistep look-ahead and exploits cheaply generated imagined data for efficient learning. Our experiments show that TRAVL can learn much faster and produce safer maneuvers compared to all the baselines.
Pixel-Wise Energy-Biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving Scenes
PDF ↗State-of-the-art (SOTA) anomaly segmentation approaches on complex urban driving scenes explore pixel-wise classification uncertainty learned from outlier exposure, or external reconstruction models. However, previous uncertainty approaches that directly associate high uncertainty to anomaly may sometimes lead to incorrect anomaly predictions, and external reconstruction models tend to be too inefficient for real-time self-driving embedded systems. In this paper, we propose a new anomaly segmentation method, named pixel-wise energy-biased abstention learning (PEBAL), that explores pixel-wise abstention learning (AL) with a model that learns an adaptive pixel-level anomaly class, and an energy-based model (EBM) that learns inlier pixel distribution. More specifically, PEBAL is based on a non-trivial joint training of EBM and AL, where EBM is trained to output high-energy for anomaly pixels (from outlier exposure) and AL is trained such that these high-energy pixels receive adaptive low penalty for being included to the anomaly class. We extensively evaluate PEBAL against the SOTA and show that it achieves the best performance across four benchmarks.
SpatialDETR: Robust Scalable Transformer-Based 3D Object Detection from Multi-View Camera Images with Global Cross-Sensor Attention
PDF ↗Based on the key idea of DETR this paper introduces an object-centric 3D object detection framework that operates on a limited number of 3D object queries instead of dense bounding box proposals followed by non-maximum suppression. After image feature extraction a decoder-only transformer architecture is trained on a set-based loss. SpatialDETR infers the classification and bounding box estimates based on attention both spatially within each image and across the different views. To fuse the multi-view information in the attention block we introduce a novel geometric positional encoding that incorporates the view ray geometry to explicitly consider the extrinsic and intrinsic camera setup. This way, the spatially-aware cross-view attention exploits arbitrary receptive fields to integrate cross-sensor data and therefore global context. Extensive experiments on the nuScenes benchmark demonstrate the potential of global attention and result in state-of-the-art performance. Code available at https://github.com/cgtuebingen/SpatialDETR.
Estimating scene flow from real-world point clouds is a fundamental task for practical 3D vision. Previous methods often rely on deep models to first extract expensive per-point features at full resolution, and then get the flow either from complex matching mechanism or feature decoding, suffering high computational cost and latency. In this work, we propose a fast hierarchical network, FH-Net, which directly gets the key points flow through a lightweight Trans-flow layer utilizing the reliable local geometry prior, and optionally back-propagates the computed sparse flows through an inverse Trans-up layer to obtain hierarchical flows at different resolutions. To focus more on challenging dynamic objects, we also provide a new copy-and-paste data augmentation technique based on dynamic object pairs generation. Moreover, to alleviate the chronic shortage of real-world training data, we establish two new large-scale datasets to this field by collecting lidar-scanned point clouds from public autonomous driving datasets and annotating the collected data through novel pseudo-labeling. Extensive experiments on both public and proposed datasets show that our method outperforms prior state-of-the-arts while running at least 7× faster at 113 FPS. Code and data are released at https://github.com/pigtigger/FH-Net.
In point cloud learning, sparsity and geometry are two core properties. Recently, many approaches have been proposed through single or multiple representations to improve the performance of point cloud semantic segmentation. However, these works fail to maintain the balance among performance, efficiency, and memory consumption, showing incapability to integrate sparsity and geometry appropriately. To address these issues, we propose the Geometry-aware Sparse Networks (GASN) by utilizing the sparsity and geometry of a point cloud in a single voxel representation. GASN mainly consists of two modules, namely Sparse Feature Encoder and Sparse Geometry Feature Enhancement. The Sparse Feature Encoder extracts the local context information, and the Sparse Geometry Feature Enhancement enhances the geometric properties of a sparse point cloud to improve both efficiency and performance. In addition, we propose deep sparse supervision in the training phase to help convergence and alleviate the memory consumption problem. Our GASN achieves state-of-the-art performance on both SemanticKITTI and Nuscenes datasets while running significantly faster and consuming less memory.
In this paper, we propose the LiDAR Distillation to bridge the domain gap induced by different LiDAR beams for 3D object detection. In many real-world applications, the LiDAR points used by mass-produced robots and vehicles usually have fewer beams than that in large-scale public datasets. Moreover, as the LiDARs are upgraded to other product models with different beam amount, it becomes challenging to utilize the labeled data captured by previous versions’ high-resolution sensors. Despite the recent progress on domain adaptive 3D detection, most methods struggle to eliminate the beam-induced domain gap. We find that it is essential to align the point cloud density of the source domain with that of the target domain during the training process. Inspired by this discovery, we propose a progressive framework to mitigate the beam-induced domain shift. In each iteration, we first generate low-beam pseudo LiDAR by downsampling the high-beam point clouds. Then the teacher-student framework is employed to distill rich information from the data with more beams. Extensive experiments on Waymo, nuScenes and KITTI datasets with three different LiDAR-based detectors demonstrate the effectiveness of our LiDAR Distillation. Notably, our approach does not increase any additional computation cost for inference. Code is available at https://github.com/weiyithu/LiDAR-Distillation.
Millimeter wave (mmWave) radars are becoming a more popular sensing modality in self-driving cars due to their favorable characteristics in adverse weather. Yet, they currently lack sufficient spatial resolution for semantic scene understanding. In this paper, we present Radatron, a system capable of accurate object detection using mmWave radar as a stand-alone sensor. To enable Radatron, we introduce a first-of-its-kind, high-resolution automotive radar dataset collected with a cascaded MIMO (Multiple Input Multiple Output) radar. Our radar achieves 5 cm range resolution and 1.2-degree angular resolution, 10× finer than other publicly available datasets. We also develop a novel hybrid radar processing and deep learning approach to achieve high vehicle detection accuracy. We train and extensively evaluate Radatron to show it achieves 92.6% AP50 and 56.3% AP75 accuracy in 2D bounding box detection, an 8% and 15.9% improvement over prior art respectively. Code and dataset are available on https://jguan.page/Radatron/.
Trajectory prediction is an essential task for successful human robot interaction, such as in autonomous driving. In this work, we address the problem of predicting future pedestrian trajectories in a first person view setting with a moving camera. To that end, we propose a novel action-based contrastive learning loss, that utilizes pedestrian action information to improve the learned trajectory embeddings. The fundamental idea behind this new loss is that trajectories of pedestrians performing the same action should be closer to each other in the feature space than the trajectories of pedestrians with significantly different actions. In other words, we argue that behavioral information about pedestrian action influences their future trajectory. Furthermore, we introduce a novel sampling strategy for trajectories that is able to effectively increase negative and positive contrastive samples. Additional synthetic trajectory samples are generated using a trained Conditional Variational Autoencoder (CVAE), which is at the core of several models developed for trajectory prediction. Results show that our proposed contrastive framework employs contextual information about pedestrian behavior, i.e. action, effectively, and it learns a better trajectory representation. Thus, integrating the proposed contrastive framework within a trajectory prediction model improves its results and outperforms state-of-the-art methods on three trajectory prediction benchmarks.
Self-supervised depth learning from monocular images normally relies on the 2D pixel-wise photometric relation between temporally adjacent image frames. However, they neither fully exploit the 3D point-wise geometric correspondences, nor effectively tackle the ambiguities in the photometric warping caused by occlusions or illumination inconsistency. To address these problems, this work proposes Density Volume Construction Network (DevNet), a novel self-supervised monocular depth learning framework, that can consider 3D spatial information, and exploit stronger geometric constraints among adjacent camera frustums. Instead of directly regressing the pixel value from a single image, our DevNet divides the camera frustum into multiple parallel planes and predicts the pointwise occlusion probability density on each plane. The final depth map is generated by integrating the density along corresponding rays. During the training process, novel regularization strategies and loss functions are introduced to mitigate photometric ambiguities and overfitting. Without obviously enlarging model parameters size or running time, DevNet outperforms several representative baselines on both the KITTI-2015 outdoor dataset and NYU-V2 indoor dataset. In particular, the root-mean-square-deviation is reduced by around 4% with DevNet on both KITTI-2015 and NYU-V2 in the task of depth estimation. Code is available at https://github.com/gitkaichenzhou/DevNet.
In this paper, we investigate the application of Vehicle-to-Everything (V2X) communication to improve the perception performance of autonomous vehicles. We present a robust cooperative perception framework with V2X communication using a novel vision Transformer. Specifically, we build a holistic attention model, namely V2X-ViT, to effectively fuse information across on-road agents (i.e., vehicles and infrastructure). V2X-ViT consists of alternating layers of heterogeneous multi-agent self-attention and multi-scale window self-attention, which captures inter-agent interaction and per-agent spatial relationships. These key modules are designed in a unified Transformer architecture to handle common V2X challenges, including asynchronous information sharing, pose errors, and heterogeneity of V2X components. To validate our approach, we create a large-scale V2X perception dataset using CARLA and OpenCDA. Extensive experimental results demonstrate that V2X-ViT sets new state-of-the-art performance for 3D object detection and achieves robust performance even under harsh noisy environments. The code is available at https://github.com/DerrickXuNu/v2x-vit.
This work addresses visual cross-view metric localization for outdoor robotics. Given a ground-level color image and a satellite patch that contains the local surroundings, the task is to identify the location of the ground camera within the satellite patch. Related work addressed this task for range-sensors (LiDAR, Radar), but for vision, only as a secondary regression step after an initial cross-view image retrieval step. Since the local satellite patch could also be retrieved through any rough localization prior (e.g. from GPS/GNSS, temporal filtering), we drop the image retrieval objective and focus on the metric localization only. We devise a novel network architecture with denser satellite descriptors, similarity matching at the bottleneck (rather than at the output as in image retrieval), and a dense spatial distribution as output to capture multi-modal localization ambiguities. We compare against a state-of-the-art regression baseline that uses global image descriptors. Quantitative and qualitative experimental results on the recently proposed VIGOR and the Oxford RobotCar datasets validate our design. The produced probabilities are correlated with localization accuracy, and can even be used to roughly estimate the ground camera’s heading when its orientation is unknown. Overall, our method reduces the median metric localization error by 51%, 37%, and 28% compared to the state-of-the-art when generalizing respectively in the same area, across areas, and across time.
Semantic segmentation of LiDAR point clouds is an important task in autonomous driving. However, training deep models via conventional supervised methods requires large datasets which are costly to label. It is critical to have label-efficient segmentation approaches to scale up the model to new operational domains or to improve performance on rare cases. While most prior works focus on indoor scenes, we are one of the first to propose a label-efficient semantic segmentation pipeline for outdoor scenes with LiDAR point clouds. Our method co-designs an efficient labeling process with semi/weakly supervised learning and is applicable to nearly any 3D semantic segmentation backbones. Specifically, we leverage geometry patterns in outdoor scenes to have a heuristic pre-segmentation to reduce the manual labeling and jointly design the learning targets with the labeling process. In the learning step, we leverage prototype learning to get more descriptive point embeddings and use multi-scan distillation to exploit richer semantics from temporally aggregated point clouds to boost the performance of single-scan models. Evaluated on the SemanticKITTI and the nuScenes datasets, we show that our proposed method outperforms existing label-efficient methods. With extremely limited human annotations (e.g., 0.1% point labels), our proposed method is even highly competitive compared to the fully supervised counterpart with 100% labels.
Master of All: Simultaneous Generalization of Urban-Scene Segmentation to All Adverse Weather Conditions
PDF ↗Computer vision systems for autonomous navigation must generalize well in adverse weather and illumination conditions expected in the real world. However, semantic segmentation of images captured in such conditions remains a challenging task for current state-of-the-art (\sota) methods trained on broad daylight images, due to the associated distribution shift. On the other hand, domain adaptation techniques developed for the purpose rely on the availability of the source data, (un)labeled target data and/or its auxiliary information (e.g., \gps). Even then, they typically adapt to a single(specific) target domain(s). To remedy this, we propose a novel, fully test time, adaptation technique, named \textit{Master of ALL} (\mall), for simultaneous generalization to multiple target domains. \mall learns to generalize on unseen adverse weather images from multiple target domains directly at the inference time. More specifically, given a pre-trained model and its parameters, \mall enforces edge consistency prior at the inference stage and updates the model based on (a) a single test sample at a time (\malls), or (b) continuously for the whole test domain (\malld). Not only the target data, \mall also does not need access to the source data and thus, can be used with any pre-trained model. Using a simple model pre-trained on daylight images, \mall outperforms specially designed adverse weather semantic segmentation methods, both in domain generalization and test-time adaptation settings. Our experiments on foggy, snow, night, cloudy, overcast, and rainy conditions demonstrate the target domain-agnostic effectiveness of our approach. We further show that \mall can improve the performance of a model on an adverse weather condition, even when the model is already pre-trained for the specific condition.
Deep learning has recently achieved significant progress in trajectory forecasting. However, the scarcity of trajectory data inhibits the data-hungry deep-learning models from learning good representations. While pre-training methods for representation learning exist in computer vision and natural language processing, they still require large-scale data. It is hard to replicate their success in trajectory forecasting due to the inadequate trajectory data (e.g., 34K samples in the nuScenes dataset). To work around the scarcity of trajectory data, we resort to another data modality closely related to trajectories--HD-maps, which are abundantly provided in existing datasets. In this paper, we propose PreTraM, a self-supervised Pre-training scheme via connecting Trajectories and Maps for trajectory forecasting. PreTraM consists of two parts: 1) Trajectory-Map Contrastive Learning, where we project trajectories and maps to a shared embedding space with cross-modal contrastive learning, 2) Map Contrastive Learning, where we enhance map representation with contrastive learning on large quantities of HD-maps. On top of popular baselines such as AgentFormer and Trajectron++, PreTraM reduces their errors by 5.5% and 6.9% relatively on the nuScenes dataset. We show that PreTraM improves data efficiency and scales well with model size.
Existing approaches for unsupervised point cloud pre-training are constrained to either scene-level or point/voxel-level instance discrimination. Scene-level methods tend to lose local details that are crucial for recognizing the road objects, while point/voxel-level methods inherently suffer from limited receptive field that is incapable of perceiving large objects or context environments. Considering region-level representations are more suitable for 3D object detection, we devise a new unsupervised point cloud pre-training framework, called ProposalContrast, that learns robust 3D representations by contrasting region proposals. Specifically, with an exhaustive set of region proposals sampled from each point cloud, geometric point relations within each proposal are modeled for creating expressive proposal representations. To better accommodate 3D detection properties, ProposalContrast optimizes with both inter-cluster and inter-proposal separation, i.e., sharpening the discriminativeness of proposal representations across semantic classes and object instances. The generalizability and transferability of ProposalContrast are verified on various 3D detectors (i.e., PV-RCNN, CenterPoint, PointPillars and PointRCNN) and datasets (i.e., KITTI, Waymo and ONCE).