Semantic Scene Research Articles

Multi-Scale Geometric Feature Extraction and Global Transformer for Real-World Indoor Point Cloud Analysis

Indoor point clouds often present significant challenges due to the complexity and variety of structures and high object similarity. The local geometric structure helps the model learn the shape features of objects at the detail level, while the global context provides overall scene semantics and spatial relationship information between objects. To address these challenges, we propose a novel network architecture, PointMSGT, which includes a multi-scale geometric feature extraction (MSGFE) module and a global Transformer (GT) module. The MSGFE module consists of a geometric feature extraction (GFE) module and a multi-scale attention (MSA) module. The GFE module reconstructs triangles through each point’s two neighbors and extracts detailed local geometric relationships by the triangle’s centroid, normal vector, and plane constant. The MSA module extracts features through multi-scale convolutions and adaptively aggregates features, focusing on both local geometric details and global semantic information at different scale levels, enhancing the understanding of complex scenes. The global Transformer employs a self-attention mechanism to capture long-range dependencies across the entire point cloud. The proposed method demonstrates competitive performance in real-world indoor scenarios, with a mIoU of 68.6% in semantic segmentation on S3DIS and OA of 86.4% in classification on ScanObjectNN.

Semantic Scene Completion in Autonomous Driving: A Two-Stream Multi-Vehicle Collaboration Approach

Vehicle-to-vehicle communication enables capturing sensor information from diverse perspectives, greatly aiding in semantic scene completion in autonomous driving. However, the misalignment of features between ego vehicle and cooperative vehicles leads to ambiguity problems, affecting accuracy and semantic information. In this paper, we propose a Two-Stream Multi-Vehicle collaboration approach (TSMV), which divides the features of collaborative vehicles into two streams and regresses interactively. To overcome the problems caused by feature misalignment, the Neighborhood Self-Cross Attention Transformer (NSCAT) module is designed to enable the ego vehicle to query the most similar local features from collaborative vehicles through cross-attention, rather than assuming spatial-temporal synchronization. A 3D occupancy map is finally generated from the features of collaborative vehicle aggregation. Experimental results on both V2VSSC and SemanticOPV2V datasets demonstrate TSMV outpace state-of-the-art collaborative semantic scene completion techniques.

SSR-2D: Semantic 3D Scene Reconstruction From 2D Images.

Most deep learning approaches to comprehensive semantic modeling of 3D indoor spaces require costly dense annotations in the 3D domain. In this work, we explore a central 3D scene modeling task, namely, semantic scene reconstruction without using any 3D annotations. The key idea of our approach is to design a trainable model that employs both incomplete 3D reconstructions and their corresponding source RGB-D images, fusing cross-domain features into volumetric embeddings to predict complete 3D geometry, color, and semantics with only 2D labeling which can be either manual or machine-generated. Our key technical innovation is to leverage differentiable rendering of color and semantics to bridge 2D observations and unknown 3D space, using the observed RGB images and 2D semantics as supervision, respectively. We additionally develop a learning pipeline and corresponding method to enable learning from imperfect predicted 2D labels, which could be additionally acquired by synthesizing in an augmented set of virtual training views complementing the original real captures, enabling more efficient self-supervision loop for semantics. As a result, our end-to-end trainable solution jointly addresses geometry completion, colorization, and semantic mapping from limited RGB-D images, without relying on any 3D ground-truth information. Our method achieves state-of-the-art performance of semantic scene completion on two large-scale benchmark datasets MatterPort3D and ScanNet, surpasses baselines even with costly 3D annotations in predicting both geometry and semantics. To our knowledge, our method is also the first 2D-driven method addressing completion and semantic segmentation of real-world 3D scans simultaneously.

CASSC: Context‐aware method for depth guided semantic scene completion

CASSC: Context‐aware method for depth guided semantic scene completion

Temporal Consistency for RGB-Thermal Data-Based Semantic Scene Understanding

Temporal Consistency for RGB-Thermal Data-Based Semantic Scene Understanding

AEFF-SSC: an attention-enhanced feature fusion for 3D semantic scene completion

Abstract Three-dimensional (3D) occupancy perception technology aims to enable autonomous vehicles to observe and understand dense 3D environments. Estimating the complete geometry and semantics of a scene solely from visual images is challenging. However, humans can easily conceive the complete form of objects based on partial key information and their own experience. This ability is crucial for recognizing and interpreting the surrounding environment. To equip 3D occupancy perception systems with a similar capability, a 3D semantic scene completion method called AEFF-SSC is proposed. This method deeply explores boundary and multi-scale information in voxels, aiming to reconstruct 3D geometry more accurately. We have specifically designed an attention-enhanced feature fusion module that effectively fuses image feature information from different scales and focuses on feature boundary information, thereby more efficiently extracting voxel features. Additionally, we introduce a semantic segmentation module driven by a 3D attention-UNet network. This module combines a 3D U-Net network with a 3D attention mechanism. Through feature fusion and feature weighting, it aids in restoring 3D spatial information and significantly improves the accuracy of segmentation results. Experimental verification on the SemanticKITTI dataset demonstrates that AEFF-SSC significantly outperforms other existing methods in terms of both geometry and semantics. Specifically, within a 12.8 m × 12.8 m area ahead, our geometric occupancy accuracy has achieved a significant improvement of 71.58%, and at the same time, the semantic segmentation accuracy has also increased remarkably by 54.20%.

Image Captioning Based on Semantic Scenes.

With the development of artificial intelligence and deep learning technologies, image captioning has become an important research direction at the intersection of computer vision and natural language processing. The purpose of image captioning is to generate corresponding natural language descriptions by understanding the content of images. This technology has broad application prospects in fields such as image retrieval, autonomous driving, and visual question answering. Currently, many researchers have proposed region-based image captioning methods. These methods generate captions by extracting features from different regions of an image. However, they often rely on local features of the image and overlook the understanding of the overall scene, leading to captions that lack coherence and accuracy when dealing with complex scenes. Additionally, image captioning methods are unable to extract complete semantic information from visual data, which may lead to captions with biases and deficiencies. Due to these reasons, existing methods struggle to generate comprehensive and accurate captions. To fill this gap, we propose the Semantic Scenes Encoder (SSE) for image captioning. It first extracts a scene graph from the image and integrates it into the encoding of the image information. Then, it extracts a semantic graph from the captions and preserves semantic information through a learnable attention mechanism, which we refer to as the dictionary. During the generation of captions, it combines the encoded information of the image and the learned semantic information to generate complete and accurate captions. To verify the effectiveness of the SSE, we tested the model on the MSCOCO dataset. The experimental results show that the SSE improves the overall quality of the captions. The improvement in scores across multiple evaluation metrics further demonstrates that the SSE possesses significant advantages when processing identical images.

2D Semantic-Guided Semantic Scene Completion

2D Semantic-Guided Semantic Scene Completion

Geometry-semantic aware for monocular 3D Semantic Scene Completion

Monocular Semantic Scene Completion (SSC) empowers intelligent devices to comprehend voxel occupancy (geometry) and semantics in 3D scenes, attracting significant attention in indoor and autonomous driving scenarios. However, existing monocular SSC models primarily map 2D images into 3D space, neglecting the potential benefits of leveraging semantic and geometric understanding in 2D. To address this, we propose the Proxy-embedding Parallel Multi-task Network (PPMNet), which aims to perceive the geometry and semantics of 3D space through depth estimation and semantic segmentation proxy tasks on the 2D perspective plane. Moreover, 2D plane features can be inversely projected into 3D space and subsequently processed using the 3D network. In addition, we enhance contextual awareness in both perspective planes and voxel grids through parallel 2D and 3D decoders. Furthermore, we employ Dual-Head Pyramid Pooling (DHPP) to aggregate information from these two representations. Finally, considering the class imbalance and label incompleteness in practical data, we design a local-to-global loss to prioritize challenging categories. Extensive experiments validate our superiority over state-of-the-art methods on the NYUv2 and SemanticKITTI datasets. The code is available at: https://github.com/luzonghao1/PPMNet.

Structural link prediction model with multi-view text semantic feature extraction

The exponential expansion of information has made text feature extraction based on simple semantic information insufficient for the multidimensional recognition of textual data. In this study, we construct a text semantic structure graph based on various perspectives and introduce weight coefficients and node clustering coefficients of co-occurrence granularity to enhance the link prediction model, in order to comprehensively capture the structural information of the text. Firstly, we jointly build the semantic structure graph based on three proposed perspectives (i.e., scene semantics, text weight, and graph structure), and propose a candidate keyword set in conjunction with an information probability retrieval model. Subsequently, we propose weight coefficients of co-occurrence granularity and node clustering coefficients to improve the link prediction model based on the semantic structure graph, enabling a more comprehensive acquisition of textual structural information. Experimental results demonstrate that our research method can reveal potential correlations and obtain more complete semantic structure information, while the WPAA evaluation index validates the effectiveness of our model.

Visual place recognition from end-to-end semantic scene text features.

We live in a visual world where text cues are abundant in urban environments. The premise for our work is for robots to capitalize on these text features for visual place recognition. A new technique is introduced that uses an end-to-end scene text detection and recognition technique to improve robot localization and mapping through Visual Place Recognition (VPR). This technique addresses several challenges such as arbitrary shaped text, illumination variation, and occlusion. The proposed model captures text strings and associated bounding boxes specifically designed for VPR tasks. The primary contribution of this work is the utilization of an end-to-end scene text spotting framework that can effectively capture irregular and occluded text in diverse environments. We conduct experimental evaluations on the Self-Collected TextPlace (SCTP) benchmark dataset, and our approach outperforms state-of-the-art methods in terms of precision and recall, which validates the effectiveness and potential of our proposed approach for VPR.

PesRec: A parametric estimation method for indoor semantic scene reconstruction from a single image

Reconstructing semantic indoor scenes is a challenging task in augmented and virtual reality. The quality of scene reconstruction is limited by the complexity and occlusion of indoor scenes. This is due to the difficulty in estimating the spatial structure of the scene and insufficient learning for object location inference. To address these challenges, we have developed PesRec, an end-to-end multi-task scene reconstruction network for parameterizing indoor semantic information. PesRec incorporates a newly designed spatial layout estimator and a 3D object detector to effectively learn scene parameter features from RGB images. We modify an object mesh generator to enhance the robustness of reconstructing indoor occluded objects through point cloud optimization in PesRec. Using the analyzed scene parameters and spatial structure, the proposed PesRec reconstructs an indoor scene by placing object meshes scaled to 3D detection boxes in an estimated layout cuboid. Extensive experiments on two benchmark datasets demonstrate that PesRec performs exceptionally well for object reconstruction with an average chamfer distance of 5.24 × 10-3 on the Pix3D dataset including 53.61 % mAP for 3D object detection and 79.7 % 3D IoU for the estimation of layout on the commonly-used SUN RGB-D datasets. The proposed computing network breaks through the limitations caused by complex indoor scenes and occlusions, showing optimization results that improve the quality of reconstruction in the fields of augmented reality and virtual reality.

Learning accurate monocular 3D voxel representation via bilateral voxel transformer

Vision-based methods for 3D scene perception have been widely explored for autonomous vehicles. However, inferring complete 3D semantic scenes from monocular 2D images is still challenging owing to the 2D-to-3D transformation. Specifically, existing methods that use Inverse Perspective Mapping (IPM) to project image features to dense 3D voxels severely suffer from the ambiguous projection problem. In this research, we present Bilateral Voxel Transformer (BVT), a novel and effective Transformer-based approach for monocular 3D semantic scene completion. BVT exploits a bilateral architecture composed of two branches for preserving the high-resolution 3D voxel representation while aggregating contexts through the proposed Tri-Axial Transformer simultaneously. To alleviate the ill-posed 2D-to-3D transformation, we adopt position-aware voxel queries and dynamically update the voxels with image features through weighted geometry-aware sampling. BVT achieves 11.8 mIoU on the challenging Semantic KITTI dataset, considerably outperforming previous works for semantic scene completion with monocular images. The code and models of BVT will be available on GitHub.

S2Former-OR: Single-Stage Bi-Modal Transformer for Scene Graph Generation in OR.

Scene graph generation (SGG) of surgical procedures is crucial in enhancing holistically cognitive intelligence in the operating room (OR). However, previous works have primarily relied on multi-stage learning, where the generated semantic scene graphs depend on intermediate processes with pose estimation and object detection. This pipeline may potentially compromise the flexibility of learning multimodal representations, consequently constraining the overall effectiveness. In this study, we introduce a novel single-stage bi-modal transformer framework for SGG in the OR, termed, S2Former-OR, aimed to complementally leverage multi-view 2D scenes and 3D point clouds for SGG in an end-to-end manner. Concretely, our model embraces a View-Sync Transfusion scheme to encourage multi-view visual information interaction. Concurrently, a Geometry-Visual Cohesion operation is designed to integrate the synergic 2D semantic features into 3D point cloud features. Moreover, based on the augmented feature, we propose a novel relation-sensitive transformer decoder that embeds dynamic entity-pair queries and relational trait priors, which enables the direct prediction of entity-pair relations for graph generation without intermediate steps. Extensive experiments have validated the superior SGG performance and lower computational cost of S2Former-OR on 4D-OR benchmark, compared with current OR-SGG methods, e.g., 3 percentage points Precision increase and 24.2M reduction in model parameters. We further compared our method with generic single-stage SGG methods with broader metrics for a comprehensive evaluation, with consistently better performance achieved. Our source code can be made available at: https://github.com/PJLallen/S2Former-OR.

TSG-Seg: Temporal-selective guidance for semi-supervised semantic segmentation of 3D LiDAR point clouds

LiDAR-based semantic scene understanding holds a pivotal role in various applications, including remote sensing and autonomous driving. However, the majority of LiDAR segmentation models rely on extensive and densely annotated training datasets, which is extremely laborious to annotate and hinder the widespread adoption of LiDAR systems. Semi-supervised learning (SSL) offers a promising solution by leveraging only a small amount of labeled data and a larger set of unlabeled data, aiming to train robust models with desired accuracy comparable to fully supervised learning. A typical pipeline of SSL involves the initial use of labeled data to train segmentation models, followed by the utilization of predictions generated from unlabeled data, which are used as pseudo-ground truths for model retraining. However, the scarcity of labeled data limits the capture of comprehensive representations, leading to the constraints of these pseudo-ground truths in reliability. We observed that objects captured by LiDAR sensors from varying perspectives showcase diverse data characteristics due to occlusions and distance variation, and LiDAR segmentation models trained with limited labels prove susceptible to these viewpoint disparities, resulting in inaccurately predicted pseudo-ground truths across viewpoints and the accumulation of retraining errors. To address this problem, we introduce the Temporal-Selective Guided Learning (TSG-Seg) framework. TSG-Seg explores temporal cues inherent in LiDAR frames to bridge the cross-viewpoint representations, fostering consistent and robust segmentation predictions across differing viewpoints. Specifically, we first establish point-wise correspondences across LiDAR frames with different time stamps through point registration. Subsequently, reliable point predictions are selected and propagated to points from adjacent views to the current view, serving as strong and refined supervision signals for subsequent model re-training to achieve better segmentation. We conducted extensive experiments on various SSL labeling setups across multiple public datasets, including SemanticKITTI and SemanticPOSS, to evaluate the effectiveness of TSG-Seg. Our results demonstrate its competitive performance and robustness in diverse scenarios, from data-limited to data-abundant settings. Notably, TSG-Seg achieves a mIoU of 48.6% using only 5% of and 62.3% with 40% of labeled data in the sequential split on SemanticKITTI. This consistently outperforms state-of-the-art segmentation methods, including GPC and LaserMix. These findings underscore TSG-Seg’s superior capability and potential for real-world applications. The project can be found at https://tsgseg.github.io.

Toward Trustworthy Automated Driving through Qualitative Scene Understanding and Explanations

<div>Understanding driving scenes and communicating automated vehicle decisions are key requirements for trustworthy automated driving. In this article, we introduce the qualitative explainable graph (QXG), which is a unified symbolic and qualitative representation for scene understanding in urban mobility. The QXG enables interpreting an automated vehicle’s environment using sensor data and machine learning models. It utilizes spatiotemporal graphs and qualitative constraints to extract scene semantics from raw sensor inputs, such as LiDAR and camera data, offering an interpretable scene model. A QXG can be incrementally constructed in real-time, making it a versatile tool for in-vehicle explanations across various sensor types. Our research showcases the potential of QXG, particularly in the context of automated driving, where it can rationalize decisions by linking the graph with observed actions. These explanations can serve diverse purposes, from informing passengers and alerting vulnerable road users to enabling post hoc analysis of prior behaviors.</div>

Depth Prior-Guided 3D Voxel Feature Fusion for 3D Semantic Estimation from Monocular Videos

Existing 3D semantic scene reconstruction methods utilize the same set of features extracted from deep learning networks for both 3D semantic estimation and geometry reconstruction, ignoring the differing requirements of semantic segmentation and geometry construction tasks. Additionally, current methods allocate 2D image features to all voxels along camera rays during the back-projection process, without accounting for empty or occluded voxels. To address these issues, we propose separating the features for 3D semantic estimation from those for 3D mesh reconstruction. We use a pretrained vision transformer network for image feature extraction and depth priors estimated by a pretrained multi-view stereo-network to guide the allocation of image features within 3D voxels during the back-projection process. The back-projected image features are aggregated within each 3D voxel via averaging, creating coherent voxel features. The resulting 3D feature volume, composed of unified voxel feature vectors, is fed into a 3D CNN with a semantic classification head to produce a 3D semantic volume. This volume can be combined with existing 3D mesh reconstruction networks to produce a 3D semantic mesh. Experimental results on real-world datasets demonstrate that the proposed method significantly increases 3D semantic estimation accuracy.

Face detection in contextual scenes.

Object and scene perception are intertwined. When objects are expected to appear within a particular scene, they are detected and categorised with greater speed and accuracy. This study examined whether such context effects also moderate the perception of social objects such as faces. Female and male faces were embedded in scenes with a stereotypical female or male context. Semantic congruency of these scene contexts influenced the categorisation of faces (Experiment 1). These effects were bi-directional, such that face sex also affected scene categorisation (Experiment 2), suggesting concurrent automatic processing of both levels. In contrast, the more elementary task of face detection was not affected by semantic scene congruency (Experiment 3), even when scenes were previewed prior to face presentation (Experiment 4). This pattern of results indicates that semantic scene context can affect categorisation of faces. However, the earlier perceptual stage of detection appears to be encapsulated from the cognitive processes that give rise to this contextual interference.

MRFTrans: Multimodal Representation Fusion Transformer for monocular 3D semantic scene completion

The complete understanding of 3D scenes is crucial in robotic visual perception, impacting tasks such as motion planning and map localization. However, due to the limited field of view and scene occlusion constraints of sensors, inferring complete scene geometry and semantic information from restricted observations is challenging. In this work, we propose a novel Multimodal Representation Fusion Transformer framework (MRFTrans) that robustly fuses semantic, geometric occupancy, and depth representations for monocular-image-based scene completion. MRFTrans centers on an affinity representation fusion transformer, integrating geometric occupancy and semantic relationships within a transformer architecture. This integration enables the modeling of long-range dependencies within scenes for inferring missing information. Additionally, we present a depth representation fusion method, efficiently extracting reliable depth knowledge from biased monocular estimates. Extensive experiments demonstrate MRFTrans’s superiority, setting a new benchmark on SemanticKITTI and NYUv2 datasets. It significantly enhances completeness and accuracy, particularly in large structures, movable objects, and scene components with major occlusions. The results underscore the benefits of the affinity-aware transformer and robust depth fusion in monocular-image-based completion.

Semantic scene segmentation for indoor autonomous vision systems: leveraging an enhanced and efficient U-NET architecture

Semantic scene segmentation for indoor autonomous vision systems: leveraging an enhanced and efficient U-NET architecture