Layer-wise correlation and attention discrepancy distillation for semantic segmentation

Chapter 12 - Semantic scene segmentation for robotics

Semantic segmentation is a fundamental task in image processing and computer vision domains. Therefore, it is a core component of most emerging industry applications such as medical imaging, autonomous driving, and agriculture. The last five years have witnessed a huge growth in the computer vision domain with the introduction of Generative Adversarial Networks (GANs) due to their ability to learn deep representations by a competitive process between two neural networks called the Generator the Discriminator. Neural Architecture Search (NAS) is undergoing a revolution aimed at automating the architectural engineering process. This paper discusses implementing a novel framework for semantic image segmentation called iSegmentor using NAS and GANs. The proposed approach is the first framework that utilizes NAS in GANs for semantic image segmentation to the best of our knowledge. The iSegmentor framework was evaluated using mean Intersection over Union (mIoU) and Mean Pixel Accuracy (MPA) metrics. The iSegmentor was mainly tested and evaluated for the autonomous driving domain. With the analysis of the past five-year literature, Cityscapes and CamVid datasets were chosen. The iSegmentor achieved 81.73 MPA and 71.90 mIoU for the Cityscapes dataset, whereas CamVid achieved 80.41 MPA and 70.63 mIoU. Further, we present future avenues that we identified with the conducted experiments to assist other researchers to extend the research area of NAS in GANs in semantic segmentation.

Panoptic Segmentation aims to provide an understanding of background (stuff) and instances of objects (things) at a pixel level. It combines the separate tasks of semantic segmentation (pixel level classification) and instance segmentation to build a single unified scene understanding task. Typically, panoptic segmentation is derived by combining semantic and instance segmentation tasks that are learned separately or jointly (multi-task networks). In general, instance segmentation networks are built by adding a foreground mask estimation layer on top of object detectors or using instance clustering methods that assign a pixel to an instance center. In this work, we present a fully convolution neural network that learns instance segmentation from semantic segmentation and instance contours (boundaries of things). Instance contours along with semantic segmentation yield a boundary aware semantic segmentation of things. Connected component labeling on these results produces instance segmentation. We merge semantic and instance segmentation results to output panoptic segmentation. We evaluate our proposed method on the CityScapes dataset to demonstrate qualitative and quantitative performances along with several ablation studies. Our overview video can be accessed from https://youtu.be/wBtcxRhG3e0.

Research on image classification sparked the latest deep-learning boom. Many downstream tasks, including semantic segmentation, benefit from it. The state-of-the-art semantic segmentation models are all based on deep learning, and they sometimes make some semantic mistakes. In a semantic segmentation dataset with a small number of categories, images are often collected from a single scene, and there is a close semantic connection between any two categories. However, in the semantic segmentation dataset collected from multiple scenes, two categories may be irrelevant. The probability of objects in one category appearing next to objects in other categories is different, which is the basis of the paper. Semantic segmentation methods need to solve two problems of positioning and classification. This paper is dedicated to correcting those clearly wrong classifications that are contrary to reality. Specifically, we first calculate the relevancy between different class pairs. Then, based on this knowledge, we infer the category of a connected component according to the relationships of this connected component with its surrounding connected components and correct the obviously wrong classifications made by a deep learning semantic segmentation model. Several well-performing deep learning models are experimented on two challenging public datasets in the field of semantic image segmentation. Our proposed method improves the performance of UPerNet, OCRNet and SETR from 40.7%, 43% and 48.64% to 42.07%, 44.09% and 49.09% mean IoU on the ADE20K validation set, and the performance of PSPNet, DeepLabV3 and OCRNet from 37.26%, 37.3% and 39.5% to 38.93%, 38.95% and 40.63% mean IoU on the COCO-Stuff dataset, which shows the effectiveness of the method.

Abstract. Semantic instance segmentation from scenes, serving as a crucial role for 3D modelling and scene understanding. Conducting semantic segmentation before grouping instances is adopted by the existing state-of-the-art methods. However, without additional refinement, semantic errors will fully propagate into the grouping stage, resulting in low overlap with the ground truth instance. Furthermore, the proposed methods focused on indoor level scenes, which are limited when directly applied to large-scale outdoor Airborne Laser Scanning (ALS) point clouds. Numerous instances, significant object density and scale variations make ALS point clouds distinct from indoor data. In order to address the problems, we proposed a geometric characterization-aware semantic instance segmentation network, which utilized both semantic and objectness score to select potential points for grouping. And in point cloud feature learning stage, hand-craft geometry features are taken as input for geometric characterization awareness. Moreover, to address errors propagated from previous modules after grouping, we have additionally designed a per-instance refinement module. To assess semantic instance segmentation, we conducted experiments on an open-source dataset. Additionally, we performed semantic segmentation experiments to evaluate the performance of our proposed point cloud feature learning method.

One promising approach for pixel-wise semantic segmentation is based on higher-order Conditional Random Fields (CKFs). We aim to selectively choose segments for the higher-order CRFs in semantic segmentation. To this end, we formulate the selection as an optimization problem. We propose three optimization criteria in relation to the selected segments, namely: a) averaged goodness, b) coverage area and c) non-overlapped area. Essentially, we desire to have best segments with maximum coverage area and maximum non-overlapped area. We apply two evolutionary optimization algorithms, namely: the genetic algorithm (GA) and the particle swarm optimization (PSO). The goodness of segments is estimated using the Latent Dirichlet Allocation approach. Experiment results show that semantic segmentation with GA-or-PSO-selected segments yields competitive semantic segmentation accuracy in comparison to that of naively using all segments. Moreover, the fewer number of segments used in semantic segmentation speeds up its computation time up to six times faster.

Semantic segmentation is an important task in self-driving cars. The aims of semantic segmentation are to recognize pre-defined objects and its pixel-by-pixel location. The most popular method in semantic segmentation is Deep learning which has considerably improved semantic image segmentation. This work does an overview for semantic segmentation using Deep learning. This works also implement comparisons in term of precision, mean IOU and processing time. Three popular algorithms are PSPNet, FCN and SegNet that are examined carefully. In detail, the aim of this work points out a trade-off between processing time and mean IOU, and also between processing time and precision. Moreover, this paper concentrates on road segmentation for embedded devices, so processing time is significantly important. This work also figures out which method is suitable for embedded devices on road segmentation.

With the emergence of deep learning, computer vision has witnessed extensive advancement and has seen immense applications in multiple domains. Specifically, image captioning has become an attractive focal direction for most machine learning experts, which includes the prerequisite of object identification, location, and semantic understanding. In this paper, semantic segmentation and image captioning are comprehensively investigated based on traditional and state‐of‐the‐art methodologies. In this survey, we deliberate on the use of deep learning techniques on the segmentation analysis of both 2D and 3D images using a fully convolutional network and other high‐level hierarchical feature extraction methods. First, each domain’s preliminaries and concept are described, and then semantic segmentation is discussed alongside its relevant features, available datasets, and evaluation criteria. Also, the semantic information capturing of objects and their attributes is presented in relation to their annotation generation. Finally, analysis of the existing methods, their contributions, and relevance are highlighted, informing the importance of these methods and illuminating a possible research continuation for the application of semantic image segmentation and image captioning approaches.

For many fundamental scene understanding tasks, it is difficult or impossible to obtain per-pixel ground truth labels from real images. We address this challenge by introducing Hypersim, a photorealistic synthetic dataset for holistic indoor scene understanding. To create our dataset, we leverage a large repository of synthetic scenes created by professional artists, and we generate 77,400 images of 461 indoor scenes with detailed per-pixel labels and corresponding ground truth geometry. Our dataset: (1) relies exclusively on publicly available 3D assets; (2) includes complete scene geometry, material information, and lighting information for every scene; (3) includes dense per-pixel semantic instance segmentations and complete camera information for every image; and (4) factors every image into diffuse reflectance, diffuse illumination, and a non-diffuse residual term that captures view-dependent lighting effects.We analyze our dataset at the level of scenes, objects, and pixels, and we analyze costs in terms of money, computation time, and annotation effort. Remarkably, we find that it is possible to generate our entire dataset from scratch, for roughly half the cost of training a popular open-source natural language processing model. We also evaluate sim-to-real transfer performance on two real-world scene understanding tasks – semantic segmentation and 3D shape prediction – where we find that pre-training on our dataset significantly improves performance on both tasks, and achieves state-of-the-art performance on the most challenging Pix3D test set. All of our rendered image data, as well as all the code we used to generate our dataset and perform our experiments, is available online.

Synthetic Aperture Radar (SAR) imagery have been one of the important tools to support earth observations and topographic measurements. It means SAR imagery are essentially rich in structures and some important target categories are difficult to recognize. Optical imagery contains rich and clear spectral information which has a good influence on semantic image segmentation. The success of deep neural networks for semantic segmentation heavily depends on large-scale and well-labeled data sets, which are hard to collect in practice. In this paper, we consider deep transfer learning for semantic segmentation, we propose a deep novel transfer learning method, which transfers a semantically segment model from SAR imagery to SAR and optical fusion imagery. The experimental results show that the method proposed achieves higher mean Intersection over Union (mIoU) with less training time compared with other methods.

Semantic image segmentation is a core task for autonomous driving, which is performed by deep models. Since training these models draws to a curse of human-based image labeling, the use of synthetic images with automatically generated labels together with unlabeled real-world images is a promising alternative. This implies addressing an unsupervised domain adaptation (UDA) problem. In this paper, we propose a new co-training procedure for synth-to-real UDA of semantic segmentation models. It performs iterations where the (unlabeled) real-world training images are labeled by intermediate deep models trained with both the (labeled) synthetic images and the real-world ones labeled in previous iterations. More specifically, a self-training stage provides two domain-adapted models and a model collaboration loop allows the mutual improvement of these two models. The final semantic segmentation labels (pseudo-labels) for the real-world images are provided by these two models. The overall procedure treats the deep models as black boxes and drives their collaboration at the level of pseudo-labeled target images, i.e., neither modifying loss functions is required, nor explicit feature alignment. We test our proposal on standard synthetic and real-world datasets for onboard semantic segmentation. Our procedure shows improvements ranging from approximately 13 to 31 mIoU points over baselines.

Accurate organ-level segmentation is essential for achieving high-throughput, non-destructive, and automated plant phenotyping. To address the challenge of intelligent acquisition of phenotypic parameters in tomato plants, we propose TSINet, an end-to-end dual-task segmentation network designed for effective and precise semantic labeling and instance recognition of tomato point clouds, based on the Pheno4D dataset. TSINet adopts an encoder–decoder architecture, where a shared encoder incorporates four Geometry-Aware Adaptive Feature Extraction Blocks (GAFEBs) to effectively capture local structures and geometric relationships in raw point clouds. Two parallel decoder branches are employed to independently decode shared high-level features for the respective segmentation tasks. Additionally, a Dual Attention-Based Feature Enhancement Module (DAFEM) is introduced to further enrich feature representations. The experimental results demonstrate that TSINet achieves superior performance in both semantic and instance segmentation, particularly excelling in challenging categories such as stems and large-scale instances. Specifically, TSINet achieves 97.00% mean precision, 96.17% recall, 96.57% F1-score, and 93.43% IoU in semantic segmentation and 81.54% mPrec, 81.69% mRec, 81.60% mCov, and 86.40% mWCov in instance segmentation. Compared with state-of-the-art methods, TSINet achieves balanced improvements across all metrics, significantly reducing false positives and false negatives while enhancing spatial completeness and segmentation accuracy. Furthermore, we conducted ablation studies and generalization tests to systematically validate the effectiveness of each TSINet component and the overall robustness of the model. This study provides an effective technological approach for high-throughput automated phenotyping of tomato plants, contributing to the advancement of intelligent agricultural management.

Semantic segmentation partitions a given image or 3D model of a scene into semantically meaning parts and assigns predetermined labels to the parts. With well‐established datasets, deep networks have been successfully used for semantic segmentation of RGB and RGB‐D images. On the other hand, due to the lack of annotated large‐scale 3D datasets, semantic segmentation for 3D scenes has not yet been much addressed with deep learning. In this paper, we present a novel framework for generating semantically segmented triangular meshes of reconstructed 3D indoor scenes using volumetric semantic fusion in the reconstruction process. Our method integrates the results of CNN‐based 2D semantic segmentation that is applied to the RGB‐D stream used for dense surface reconstruction. To reduce the artifacts from noise and uncertainty of single‐view semantic segmentation, we introduce adaptive integration for the volumetric semantic fusion and CRF‐based semantic label regularization. With these methods, our framework can easily generate a high‐quality triangular mesh of the reconstructed 3D scene with dense (i.e., per‐vertex) semantic labels. Extensive experiments demonstrate that our semantic segmentation results of 3D scenes achieves the state‐of‐the‐art performance compared to the previous voxel‐based and point cloud‐based methods.

Crowdsourced images hold information could potentially be used to remotely monitor heritage sites, and reduce human and capital resources devoted to on-site inspections. This article proposes a combination of semantic image segmentation and photogrammetry to monitor changes in built heritage sites. In particular, this article focuses on segmenting potentially damaging plants from the surrounding stone masonry and other image elements. The method compares different backend models and two model architectures: (i) a one-stage model that segments seven classes within the image, and (ii) a two-stage model that uses the results from the first stage to refine a binary segmentation for the plant class. The final selected model can achieve an overall IoU of 66.9% for seven classes (54.6% for one-stage plant, 56.2% for two-stage plant). Further, the segmentation output is combined with photogrammetry to build a 3D segmented model to measure the area of biological growth. Lastly, the main findings from this paper are: (i) With the help of transfer learning and proper choice of model architecture, image segmentation can be easily applied to analyze crowdsourcing data. (ii) Photogrammetry can be combined with image segmentation to alleviate image distortions for monitoring purpose. (iii) Beyond the measurement of plant area, this method has the potential to be easily transferred into other tasks, such as monitoring cracks and erosion, or as a masking tool in the photogrammetry workflow.

In order to enhance the image understanding of different regions for national costume grayscale image automatic colorization, let coloring tasks take advantage of semantic conditions, also let it can apply the human parsing semantic segmentation method to the national costume grayscale image for semantic segmentation task. This paper proposes a semantic segmentation model for context embedding based on edge perceiving. Aiming at the features of national costume grayscale image, more optimizing the model and loss function. The national costume grayscale image semantic segmentation is different from semantic segmentation of the color image, this task is more difficult for the grayscale image has no color feature. In this paper, edge information and edge consistency constraints are used to improve the national costume grayscale image coloring effect. The experimental results show that the model designed in this paper can obtain more accurate fine-grained semantic segmentation results for the national costume grayscale image.