We describe a new general method for segmentation in MRI scans using Topological Data Analysis (TDA), offering several advantages over traditional machine learning approaches. It works in three steps, first identifying the whole object to segment via automatic thresholding, then detecting a distinctive subset whose topology is known in advance, and finally deducing the various components of the segmentation. Although convoking classical ideas of TDA, such an algorithm has never been proposed separately from deep learning methods. To achieve this, our approach takes into account, in addition to the homology of the image, the localization of representative cycles, a piece of information that seems never to have been exploited in this context. In particular, it offers the ability to perform segmentation without the need for large annotated data sets. TDA also provides a more interpretable and stable framework for segmentation by explicitly mapping topological features to segmentation components. By adapting the geometric object to be detected, the algorithm can be adjusted to a wide range of data segmentation challenges. We carefully study the examples of glioblastoma segmentation in brain MRI, where a sphere is to be detected, as well as myocardium in cardiac MRI, involving a cylinder, and cortical plate detection in fetal brain MRI, whose 2D slices are circles. We compare our method to state-of-the-art algorithms.
3D point cloud understanding is an important component in autonomous driving and robotics. In this paper, we present a novel Embedding-Querying paradigm (EQ- Paradigm) for 3D understanding tasks including detection, segmentation, and classification. EQ-Paradigm is a unified paradigm that enables the combination of any existing 3D backbone architectures with different task heads. Under the EQ-Paradigm, the input is firstly encoded in the embedding stage with an arbitrary feature extraction architecture, which is independent of tasks and heads. Then, the querying stage enables the encoded features to be applicable for diverse task heads. This is achieved by introducing an intermediate representation, i.e., Q-representation, in the querying stage to serve as a bridge between the embedding stage and task heads. We design a novel Q- Net as the querying stage network. Extensive experimental results on various 3D tasks, including object detection, semantic segmentation and shape classification, show that EQ-Paradigm in tandem with Q-Net is a general and effective pipeline, which enables a flexible collaboration of backbones and heads, and further boosts the performance of the state-of-the-art methods. Codes and models are available at https://github.com/dvlab-research/DeepVision3D.
Occlusal contacts are the locations at which the occluding surfaces of the maxilla and the mandible posterior teeth meet. Occlusal contact detection is a vital tool for restoring the loss of masticatory function and is a mandatory assessment in the field of dentistry, with particular importance in prosthodontics and restorative dentistry. The most common method for occlusal contact detection is articulating paper. However, this method can indicate significant medically false positive and medically false negative contact areas, leaving the identification of true occlusal indications to clinicians. To address this, we propose a multiclass Vision Transformer and Fully Convolutional Network ensemble semantic segmentation model with a combination hierarchical loss function, which we name as Hierarchical Fully Convolutional Branch Transformer (H-FCBFormer). We also propose a method of generating medically true positive semantic segmentation masks derived from expert annotated articulating paper masks and gold standard masks. The proposed model outperforms other machine learning methods evaluated at detecting medically true positive contacts and performs better than dentists in terms of accurately identifying object-wise occlusal contact areas while taking significantly less time to identify them. Code is available at https://github.com/Banksylel/H-FCBFormer.
Single Image Super-Resolution (SISR) is a vital technique for improving the visual quality of low-resolution images. While recent deep learning models have made significant advancements in SISR, they often encounter computational challenges that hinder their deployment in resource-limited or time-sensitive environments. To overcome these issues, we present Sebica, a lightweight network that incorporates spatial and efficient bidirectional channel attention mechanisms. Sebica significantly reduces computational costs while maintaining high reconstruction quality, achieving PSNR/SSIM scores of 28.29/0.7976 and 30.18/0.8330 on the Div2K and Flickr2K datasets, respectively. These results surpass most baseline lightweight models and are comparable to the highest-performing model, but with only 17% and 15% of the parameters and GFLOPs. Additionally, our small version of Sebica has only 7.9K parameters and 0.41 GFLOPS, representing just 3% of the parameters and GFLOPs of the highest-performing model, while still achieving PSNR and SSIM metrics of 28.12/0.7931 and 0.3009/0.8317, on the Flickr2K dataset respectively. In addition, Sebica demonstrates significant improvements in real-world applications, specifically in object detection tasks, where it enhances detection accuracy in traffic video scenarios.
Grasping objects is a fundamental yet important capability of robots, and many tasks such as sorting and picking rely on this skill. The prerequisite for stable grasping is the ability to correctly identify suitable grasping positions. However, finding appropriate grasping points is challenging due to the diverse shapes, varying density distributions, and significant differences between the barycenter of various objects. In the past few years, researchers have proposed many methods to address the above-mentioned issues and achieved very good results on publicly available datasets such as the Cornell dataset and the Jacquard dataset. The problem is that the backgrounds of Cornell and Jacquard datasets are relatively simple - typically just a whiteboard, while in real-world operational environments, the background could be complex and noisy. Moreover, in real-world scenarios, robots usually only need to grasp fixed types of objects. To address the aforementioned issues, we proposed a large-scale grasp detection dataset called NBMOD: Noisy Background Multi-Object Dataset for grasp detection, which consists of 31,500 RGB-D images of 20 different types of fruits. Accurate prediction of angles has always been a challenging problem in the detection task of oriented bounding boxes. This paper presents a Rotation Anchor Mechanism (RAM) to address this issue. Considering the high real-time requirement of robotic systems, we propose a series of lightweight architectures called RA-GraspNet (GraspNet with Rotation Anchor): RARA (network with Rotation Anchor and Region Attention), RAST (network with Rotation Anchor and Semi Transformer), and RAGT (network with Rotation Anchor and Global Transformer) to tackle this problem. Among them, the RAGT-3/3 model achieves an accuracy of 99% on the NBMOD dataset. The NBMOD and our code are available at https://github.com/kmittle/Grasp-Detection-NBMOD.
In this paper, we propose PointSeg, a real-time end-to-end semantic segmentation method for road-objects based on spherical images. We take the spherical image, which is transformed from the 3D LiDAR point clouds, as input of the convolutional neural networks (CNNs) to predict the point-wise semantic map. To make PointSeg applicable on a mobile system, we build the model based on the light-weight network, SqueezeNet, with several improvements. It maintains a good balance between memory cost and prediction performance. Our model is trained on spherical images and label masks projected from the KITTI 3D object detection dataset. Experiments show that PointSeg can achieve competitive accuracy with 90fps on a single GPU 1080ti. which makes it quite compatible for autonomous driving applications.
We propose a novel data augmentation method `GridMask' in this paper. It utilizes information removal to achieve state-of-the-art results in a variety of computer vision tasks. We analyze the requirement of information dropping. Then we show limitation of existing information dropping algorithms and propose our structured method, which is simple and yet very effective. It is based on the deletion of regions of the input image. Our extensive experiments show that our method outperforms the latest AutoAugment, which is way more computationally expensive due to the use of reinforcement learning to find the best policies. On the ImageNet dataset for recognition, COCO2017 object detection, and on Cityscapes dataset for semantic segmentation, our method all notably improves performance over baselines. The extensive experiments manifest the effectiveness and generality of the new method.
In vision-based object classification systems imaging sensors perceive the environment and machine learning is then used to detect and classify objects for decision-making purposes; e.g., to maneuver an automated vehicle around an obstacle or to raise an alarm to indicate the presence of an intruder in surveillance settings. In this work we demonstrate how the perception domain can be remotely and unobtrusively exploited to enable an attacker to create spurious objects or alter an existing object. An automated system relying on a detection/classification framework subject to our attack could be made to undertake actions with catastrophic results due to attacker-induced misperception. We focus on camera-based systems and show that it is possible to remotely project adversarial patterns into camera systems by exploiting two common effects in optical imaging systems, viz., lens flare/ghost effects and auto-exposure control. To improve the robustness of the attack to channel effects, we generate optimal patterns by integrating adversarial machine learning techniques with a trained end-to-end channel model. We experimentally demonstrate our attacks using a low-cost projector, on three different image datasets, in indoor and outdoor environments, and with three different cameras. Experimental results show that, depending on the projector-camera distance, attack success rates can reach as high as 100% and under targeted conditions.
Mixup-based augmentation has been found to be effective for generalizing models during training, especially for Vision Transformers (ViTs) since they can easily overfit. However, previous mixup-based methods have an underlying prior knowledge that the linearly interpolated ratio of targets should be kept the same as the ratio proposed in input interpolation. This may lead to a strange phenomenon that sometimes there is no valid object in the mixed image due to the random process in augmentation but there is still response in the label space. To bridge such gap between the input and label spaces, we propose TransMix, which mixes labels based on the attention maps of Vision Transformers. The confidence of the label will be larger if the corresponding input image is weighted higher by the attention map. TransMix is embarrassingly simple and can be implemented in just a few lines of code without introducing any extra parameters and FLOPs to ViT-based models. Experimental results show that our method can consistently improve various ViT-based models at scales on ImageNet classification. After pre-trained with TransMix on ImageNet, the ViT-based models also demonstrate better transferability to semantic segmentation, object detection and instance segmentation. TransMix also exhibits to be more robust when evaluating on 4 different benchmarks. Code will be made publicly available at https://github.com/Beckschen/TransMix.
Recent works on open-vocabulary 3D instance segmentation show strong promise, but at the cost of slow inference speed and high computation requirements. This high computation cost is typically due to their heavy reliance on 3D clip features, which require computationally expensive 2D foundation models like Segment Anything (SAM) and CLIP for multi-view aggregation into 3D. As a consequence, this hampers their applicability in many real-world applications that require both fast and accurate predictions. To this end, we propose a fast yet accurate open-vocabulary 3D instance segmentation approach, named Open-YOLO 3D, that effectively leverages only 2D object detection from multi-view RGB images for open-vocabulary 3D instance segmentation. We address this task by generating class-agnostic 3D masks for objects in the scene and associating them with text prompts. We observe that the projection of class-agnostic 3D point cloud instances already holds instance information; thus, using SAM might only result in redundancy that unnecessarily increases the inference time. We empirically find that a better performance of matching text prompts to 3D masks can be achieved in a faster fashion with a 2D object detector. We validate our Open-YOLO 3D on two benchmarks, ScanNet200 and Replica, under two scenarios: (i) with ground truth masks, where labels are required for given object proposals, and (ii) with class-agnostic 3D proposals generated from a 3D proposal network. Our Open-YOLO 3D achieves state-of-the-art performance on both datasets while obtaining up to $\sim$16$\times$ speedup compared to the best existing method in literature. On ScanNet200 val. set, our Open-YOLO 3D achieves mean average precision (mAP) of 24.7\% while operating at 22 seconds per scene. Code and model are available at github.com/aminebdj/OpenYOLO3D.
Long-tail distribution is widely spread in real-world applications. Due to the extremely small ratio of instances, tail categories often show inferior accuracy. In this paper, we find such performance bottleneck is mainly caused by the imbalanced gradients, which can be categorized into two parts: (1) positive part, deriving from the samples of the same category, and (2) negative part, contributed by other categories. Based on comprehensive experiments, it is also observed that the gradient ratio of accumulated positives to negatives is a good indicator to measure how balanced a category is trained. Inspired by this, we come up with a gradient-driven training mechanism to tackle the long-tail problem: re-balancing the positive/negative gradients dynamically according to current accumulative gradients, with a unified goal of achieving balance gradient ratios. Taking advantage of the simple and flexible gradient mechanism, we introduce a new family of gradient-driven loss functions, namely equalization losses. We conduct extensive experiments on a wide spectrum of visual tasks, including two-stage/single-stage long-tailed object detection (LVIS), long-tailed image classification (ImageNet-LT, Places-LT, iNaturalist), and long-tailed semantic segmentation (ADE20K). Our method consistently outperforms the baseline models, demonstrating the effectiveness and generalization ability of the proposed equalization losses. Codes will be released at https://github.com/ModelTC/United-Perception.
Recent work has focused on generating synthetic imagery to increase the size and variability of training data for learning visual tasks in urban scenes. This includes increasing the occurrence of occlusions or varying environmental and weather effects. However, few have addressed modeling variation in the sensor domain. Sensor effects can degrade real images, limiting generalizability of network performance on visual tasks trained on synthetic data and tested in real environments. This paper proposes an efficient, automatic, physically-based augmentation pipeline to vary sensor effects --chromatic aberration, blur, exposure, noise, and color cast-- for synthetic imagery. In particular, this paper illustrates that augmenting synthetic training datasets with the proposed pipeline reduces the domain gap between synthetic and real domains for the task of object detection in urban driving scenes.
Current methods for Earth observation tasks such as semantic mapping, map alignment, and change detection rely on near-nadir images; however, often the first available images in response to dynamic world events such as natural disasters are oblique. These tasks are much more difficult for oblique images due to observed object parallax. There has been recent success in learning to regress geocentric pose, defined as height above ground and orientation with respect to gravity, by training with airborne lidar registered to satellite images. We present a model for this novel task that exploits affine invariance properties to outperform state of the art performance by a wide margin. We also address practical issues required to deploy this method in the wild for real-world applications. Our data and code are publicly available.
Facial Emotion Recognition is an inherently difficult problem, due to vast differences in facial structures of individuals and ambiguity in the emotion displayed by a person. Recently, a lot of work is being done in the field of Facial Emotion Recognition, and the performance of the CNNs for this task has been inferior compared to the results achieved by CNNs in other fields like Object detection, Facial recognition etc. In this paper, we propose a multi-task learning algorithm, in which a single CNN detects gender, age and race of the subject along with their emotion. We validate this proposed methodology using two datasets containing real-world images. The results show that this approach is significantly better than the current State of the art algorithms for this task.
This paper does not attempt to design a state-of-the-art method for visual recognition but investigates a more efficient way to make use of convolutions to encode spatial features. By comparing the design principles of the recent convolutional neural networks ConvNets) and Vision Transformers, we propose to simplify the self-attention by leveraging a convolutional modulation operation. We show that such a simple approach can better take advantage of the large kernels (>=7x7) nested in convolutional layers. We build a family of hierarchical ConvNets using the proposed convolutional modulation, termed Conv2Former. Our network is simple and easy to follow. Experiments show that our Conv2Former outperforms existent popular ConvNets and vision Transformers, like Swin Transformer and ConvNeXt in all ImageNet classification, COCO object detection and ADE20k semantic segmentation.
We present methods for conditional and residual coding in the context of scalable coding for humans and machines. Our focus is on optimizing the rate-distortion performance of the reconstruction task using the information available in the computer vision task. We include an information analysis of both approaches to provide baselines and also propose an entropy model suitable for conditional coding with increased modelling capacity and similar tractability as previous work. We apply these methods to image reconstruction, using, in one instance, representations created for semantic segmentation on the Cityscapes dataset, and in another instance, representations created for object detection on the COCO dataset. In both experiments, we obtain similar performance between the conditional and residual methods, with the resulting rate-distortion curves contained within our baselines.
Diplomatics, the analysis of medieval charters, is a major field of research in which paleography is applied. Annotating data, if performed by laymen, needs validation and correction by experts. In this paper, we propose an effective and efficient annotation approach for charter segmentation, essentially reducing it to object detection. This approach allows for a much more efficient use of the paleographer's time and produces results that can compete and even outperform pixel-level segmentation in some use cases. Further experiments shed light on how to design a class ontology in order to make the best use of annotators' time and effort. Exploiting the presence of calibration cards in the image, we further annotate the data with the physical length in pixels and train regression neural networks to predict it from image patches.
The current main stream methods formulate their video saliency mainly from two independent venues, i.e., the spatial and temporal branches. As a complementary component, the main task for the temporal branch is to intermittently focus the spatial branch on those regions with salient movements. In this way, even though the overall video saliency quality is heavily dependent on its spatial branch, however, the performance of the temporal branch still matter. Thus, the key factor to improve the overall video saliency is how to further boost the performance of these branches efficiently. In this paper, we propose a novel spatiotemporal network to achieve such improvement in a full interactive fashion. We integrate a lightweight temporal model into the spatial branch to coarsely locate those spatially salient regions which are correlated with trustworthy salient movements. Meanwhile, the spatial branch itself is able to recurrently refine the temporal model in a multi-scale manner. In this way, both the spatial and temporal branches are able to interact with each other, achieving the mutual performance improvement. Our method is easy to implement yet effective, achieving high quality video saliency detection in real-time speed with 50 FPS.
Non-nutritive sucking (NNS), which refers to the act of sucking on a pacifier, finger, or similar object without nutrient intake, plays a crucial role in assessing healthy early development. In the case of preterm infants, NNS behavior is a key component in determining their readiness for feeding. In older infants, the characteristics of NNS behavior offer valuable insights into neural and motor development. Additionally, NNS activity has been proposed as a potential safeguard against sudden infant death syndrome (SIDS). However, the clinical application of NNS assessment is currently hindered by labor-intensive and subjective finger-in-mouth evaluations. Consequently, researchers often resort to expensive pressure transducers for objective NNS signal measurement. To enhance the accessibility and reliability of NNS signal monitoring for both clinicians and researchers, we introduce a vision-based algorithm designed for non-contact detection of NNS activity using baby monitor footage in natural settings. Our approach involves a comprehensive exploration of optical flow and temporal convolutional networks, enabling the detection and amplification of subtle infant-sucking signals. We successfully classify short video clips of uniform length into NNS and non-NNS periods. Furthermore, we investigate manual and learning-based techniques to piece together local classification results, facilitating the segmentation of longer mixed-activity videos into NNS and non-NNS segments of varying duration. Our research introduces two novel datasets of annotated infant videos, including one sourced from our clinical study featuring 19 infant subjects and 183 hours of overnight baby monitor footage.
Smart focal-plane and in-chip image processing has emerged as a crucial technology for vision-enabled embedded systems with energy efficiency and privacy. However, the lack of special datasets providing examples of the data that these neuromorphic sensors compute to convey visual information has hindered the adoption of these promising technologies. Neuromorphic imager variants, including event-based sensors, produce various representations such as streams of pixel addresses representing time and locations of intensity changes in the focal plane, temporal-difference data, data sifted/thresholded by temporal differences, image data after applying spatial transformations, optical flow data, and/or statistical representations. To address the critical barrier to entry, we provide an annotated, temporal-threshold-based vision dataset specifically designed for face detection tasks derived from the same videos used for Aff-Wild2. By offering multiple threshold levels (e.g., 4, 8, 12, and 16), this dataset allows for comprehensive evaluation and optimization of state-of-the-art neural architectures under varying conditions and settings compared to traditional methods. The accompanying tool flow for generating event data from raw videos further enhances accessibility and usability. We anticipate that this resource will significantly support the development of robust vision systems based on smart sensors that can process based on temporal-difference thresholds, enabling more accurate and efficient object detection and localization and ultimately promoting the broader adoption of low-power, neuromorphic imaging technologies. To support further research, we publicly released the dataset at \url{https://dx.doi.org/10.21227/bw2e-dj78}.