Human perception reliably identifies movable and immovable parts of 3D scenes, and completes the 3D structure of objects and background from incomplete observations. We learn this skill not via labeled examples, but simply by observing objects move. In this work, we propose an approach that observes unlabeled multi-view videos at training time and learns to map a single image observation of a complex scene, such as a street with cars, to a 3D neural scene representation that is disentangled into movable and immovable parts while plausibly completing its 3D structure. We separately parameterize movable and immovable scene parts via 2D neural ground plans. These ground plans are 2D grids of features aligned with the ground plane that can be locally decoded into 3D neural radiance fields. Our model is trained self-supervised via neural rendering. We demonstrate that the structure inherent to our disentangled 3D representation enables a variety of downstream tasks in street-scale 3D scenes using simple heuristics, such as extraction of object-centric 3D representations, novel view synthesis, instance segmentation, and 3D bounding box prediction, highlighting its value as a backbone for data-efficient 3D scene understanding models. This disentanglement further enables scene editing via object manipulation such as deletion, insertion, and rigid-body motion.
相關內容
Neural volumetric representations have shown the potential that MLP networks can be trained with multi-view calibrated images to represent scene geometry and appearance, without explicit 3D supervision. Object segmentation can enrich many downstream applications based on the learned radiance field. However, introducing hand-crafted segmentation to define regions of interest in a complex real-world scene are non-trivial and expensive as it acquires per view annotation. This paper carries out the exploration of self-supervised learning for object segmentation using NeRF for complex real-world scenes. Our framework, NeRF-SOS, couples object segmentation and neural radiance field to segment objects in any view within a scene. By proposing a novel collaborative contrastive loss in both appearance and geometry levels, NeRF-SOS encourages NeRF models to distill compact geometry-aware segmentation clusters from their density fields and the self-supervised pre-trained 2D visual features. The self-supervised object segmentation framework can be applied to various NeRF models that both lead to photo-realistic rendering results and convincing segmentations for both indoor and outdoor scenarios. Extensive results on the LLFF, Tank and Temple datasets validate the effectiveness of NeRF-SOS. It consistently surpasses other image-based self-supervised baselines and even captures finer details than supervised Semantic-NeRF.
Semantic segmentation is a challenging computer vision task demanding a significant amount of pixel-level annotated data. Producing such data is a time-consuming and costly process, especially for domains with a scarcity of experts, such as medicine or forensic anthropology. While numerous semi-supervised approaches have been developed to make the most from the limited labeled data and ample amount of unlabeled data, domain-specific real-world datasets often have characteristics that both reduce the effectiveness of off-the-shelf state-of-the-art methods and also provide opportunities to create new methods that exploit these characteristics. We propose and evaluate a semi-supervised method that reuses available labels for unlabeled images of a dataset by exploiting existing similarities, while dynamically weighting the impact of these reused labels in the training process. We evaluate our method on a large dataset of human decomposition images and find that our method, while conceptually simple, outperforms state-of-the-art consistency and pseudo-labeling-based methods for the segmentation of this dataset. This paper includes graphic content of human decomposition.
Detecting dangerous traffic agents in videos captured by vehicle-mounted dashboard cameras (dashcams) is essential to facilitate safe navigation in a complex environment. Accident-related videos are just a minor portion of the driving video big data, and the transient pre-accident processes are highly dynamic and complex. Besides, risky and non-risky traffic agents can be similar in their appearance. These make risky object localization in the driving video particularly challenging. To this end, this paper proposes an attention-guided multistream feature fusion network (AM-Net) to localize dangerous traffic agents from dashcam videos. Two Gated Recurrent Unit (GRU) networks use object bounding box and optical flow features extracted from consecutive video frames to capture spatio-temporal cues for distinguishing dangerous traffic agents. An attention module coupled with the GRUs learns to attend to the traffic agents relevant to an accident. Fusing the two streams of features, AM-Net predicts the riskiness scores of traffic agents in the video. In supporting this study, the paper also introduces a benchmark dataset called Risky Object Localization (ROL). The dataset contains spatial, temporal, and categorical annotations with the accident, object, and scene-level attributes. The proposed AM-Net achieves a promising performance of 85.73% AUC on the ROL dataset. Meanwhile, the AM-Net outperforms current state-of-the-art for video anomaly detection by 6.3% AUC on the DoTA dataset. A thorough ablation study further reveals AM-Net's merits by evaluating the contributions of its different components.
Personalised 3D vascular models are valuable for diagnosis, prognosis and treatment planning in patients with cardiovascular disease. Traditionally, such models have been constructed with explicit representations such as meshes and voxel masks, or implicit representations such as radial basis functions or atomic (tubular) shapes. Here, we propose to represent surfaces by the zero level set of their signed distance function (SDF) in a differentiable implicit neural representation (INR). This allows us to model complex vascular structures with a representation that is implicit, continuous, light-weight, and easy to integrate with deep learning algorithms. We here demonstrate the potential of this approach with three practical examples. First, we obtain an accurate and watertight surface for an abdominal aortic aneurysm (AAA) from CT images and show robust fitting from as little as 200 points on the surface. Second, we simultaneously fit nested vessel walls in a single INR without intersections. Third, we show how 3D models of individual arteries can be smoothly blended into a single watertight surface. Our results show that INRs are a flexible representation with potential for minimally interactive annotation and manipulation of complex vascular structures.
Weakly supervised point cloud semantic segmentation methods that require 1\% or fewer labels, hoping to realize almost the same performance as fully supervised approaches, which recently, have attracted extensive research attention. A typical solution in this framework is to use self-training or pseudo labeling to mine the supervision from the point cloud itself, but ignore the critical information from images. In fact, cameras widely exist in LiDAR scenarios and this complementary information seems to be greatly important for 3D applications. In this paper, we propose a novel cross-modality weakly supervised method for 3D segmentation, incorporating complementary information from unlabeled images. Basically, we design a dual-branch network equipped with an active labeling strategy, to maximize the power of tiny parts of labels and directly realize 2D-to-3D knowledge transfer. Afterwards, we establish a cross-modal self-training framework in an Expectation-Maximum (EM) perspective, which iterates between pseudo labels estimation and parameters updating. In the M-Step, we propose a cross-modal association learning to mine complementary supervision from images by reinforcing the cycle-consistency between 3D points and 2D superpixels. In the E-step, a pseudo label self-rectification mechanism is derived to filter noise labels thus providing more accurate labels for the networks to get fully trained. The extensive experimental results demonstrate that our method even outperforms the state-of-the-art fully supervised competitors with less than 1\% actively selected annotations.
Creating novel views from a single image has achieved tremendous strides with advanced autoregressive models. Although recent methods generate high-quality novel views, synthesizing with only one explicit or implicit 3D geometry has a trade-off between two objectives that we call the ``seesaw'' problem: 1) preserving reprojected contents and 2) completing realistic out-of-view regions. Also, autoregressive models require a considerable computational cost. In this paper, we propose a single-image view synthesis framework for mitigating the seesaw problem. The proposed model is an efficient non-autoregressive model with implicit and explicit renderers. Motivated by characteristics that explicit methods well preserve reprojected pixels and implicit methods complete realistic out-of-view region, we introduce a loss function to complement two renderers. Our loss function promotes that explicit features improve the reprojected area of implicit features and implicit features improve the out-of-view area of explicit features. With the proposed architecture and loss function, we can alleviate the seesaw problem, outperforming autoregressive-based state-of-the-art methods and generating an image $\approx$100 times faster. We validate the efficiency and effectiveness of our method with experiments on RealEstate10K and ACID datasets.
Self-supervised monocular depth estimation, aiming to learn scene depths from single images in a self-supervised manner, has received much attention recently. In spite of recent efforts in this field, how to learn accurate scene depths and alleviate the negative influence of occlusions for self-supervised depth estimation, still remains an open problem. Addressing this problem, we firstly empirically analyze the effects of both the continuous and discrete depth constraints which are widely used in the training process of many existing works. Then inspired by the above empirical analysis, we propose a novel network to learn an Occlusion-aware Coarse-to-Fine Depth map for self-supervised monocular depth estimation, called OCFD-Net. Given an arbitrary training set of stereo image pairs, the proposed OCFD-Net does not only employ a discrete depth constraint for learning a coarse-level depth map, but also employ a continuous depth constraint for learning a scene depth residual, resulting in a fine-level depth map. In addition, an occlusion-aware module is designed under the proposed OCFD-Net, which is able to improve the capability of the learnt fine-level depth map for handling occlusions. Experimental results on KITTI demonstrate that the proposed method outperforms the comparative state-of-the-art methods under seven commonly used metrics in most cases. In addition, experimental results on Make3D demonstrate the effectiveness of the proposed method in terms of the cross-dataset generalization ability under four commonly used metrics. The code is available at //github.com/ZM-Zhou/OCFD-Net_pytorch.
Since the preparation of labeled data for training semantic segmentation networks of point clouds is a time-consuming process, weakly supervised approaches have been introduced to learn from only a small fraction of data. These methods are typically based on learning with contrastive losses while automatically deriving per-point pseudo-labels from a sparse set of user-annotated labels. In this paper, our key observation is that the selection of what samples to annotate is as important as how these samples are used for training. Thus, we introduce a method for weakly supervised segmentation of 3D scenes that combines self-training with active learning. The active learning selects points for annotation that likely result in performance improvements to the trained model, while the self-training makes efficient use of the user-provided labels for learning the model. We demonstrate that our approach leads to an effective method that provides improvements in scene segmentation over previous works and baselines, while requiring only a small number of user annotations.
Most object recognition approaches predominantly focus on learning discriminative visual patterns while overlooking the holistic object structure. Though important, structure modeling usually requires significant manual annotations and therefore is labor-intensive. In this paper, we propose to "look into object" (explicitly yet intrinsically model the object structure) through incorporating self-supervisions into the traditional framework. We show the recognition backbone can be substantially enhanced for more robust representation learning, without any cost of extra annotation and inference speed. Specifically, we first propose an object-extent learning module for localizing the object according to the visual patterns shared among the instances in the same category. We then design a spatial context learning module for modeling the internal structures of the object, through predicting the relative positions within the extent. These two modules can be easily plugged into any backbone networks during training and detached at inference time. Extensive experiments show that our look-into-object approach (LIO) achieves large performance gain on a number of benchmarks, including generic object recognition (ImageNet) and fine-grained object recognition tasks (CUB, Cars, Aircraft). We also show that this learning paradigm is highly generalizable to other tasks such as object detection and segmentation (MS COCO). Project page: //github.com/JDAI-CV/LIO.
Semantic reconstruction of indoor scenes refers to both scene understanding and object reconstruction. Existing works either address one part of this problem or focus on independent objects. In this paper, we bridge the gap between understanding and reconstruction, and propose an end-to-end solution to jointly reconstruct room layout, object bounding boxes and meshes from a single image. Instead of separately resolving scene understanding and object reconstruction, our method builds upon a holistic scene context and proposes a coarse-to-fine hierarchy with three components: 1. room layout with camera pose; 2. 3D object bounding boxes; 3. object meshes. We argue that understanding the context of each component can assist the task of parsing the others, which enables joint understanding and reconstruction. The experiments on the SUN RGB-D and Pix3D datasets demonstrate that our method consistently outperforms existing methods in indoor layout estimation, 3D object detection and mesh reconstruction.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191