Small object detection for 3D point cloud is a challenging problem because of two limitations: (1) Perceiving small objects is much more diffcult than normal objects due to the lack of valid points. (2) Small objects are easily blocked which breaks the shape of their meshes in 3D point cloud. In this paper, we propose a pillar set abstraction (PSA) and foreground point compensation (FPC) and design a point-based detection network, PSA-Det3D, to improve the detection performance for small object. The PSA embeds a pillar query operation on the basis of set abstraction (SA) to expand its receptive field of the network, which can aggregate point-wise features effectively. To locate more occluded objects, we persent a proposal generation layer consisting of a foreground point segmentation and a FPC module. Both the foreground points and the estimated centers are finally fused together to generate the detection result. The experiments on the KITTI 3D detection benchmark show that our proposed PSA-Det3D outperforms other algorithms with high accuracy for small object detection.
We tackle the problem of novel class discovery and localization (NCDL). In this setting, we assume a source dataset with supervision for only some object classes. Instances of other classes need to be discovered, classified, and localized automatically based on visual similarity without any human supervision. To tackle NCDL, we propose a two-stage object detection network Region-based NCDL (RNCDL) that uses a region proposal network to localize regions of interest (RoIs). We then train our network to learn to classify each RoI, either as one of the known classes, seen in the source dataset, or one of the novel classes, with a long-tail distribution constraint on the class assignments, reflecting the natural frequency of classes in the real world. By training our detection network with this objective in an end-to-end manner, it learns to classify all region proposals for a large variety of classes, including those not part of the labeled object class vocabulary. Our experiments conducted using COCO and LVIS datasets reveal that our method is significantly more effective than multi-stage pipelines that rely on traditional clustering algorithms. Furthermore, we demonstrate the generality of our approach by applying our method to a large-scale Visual Genome dataset, where our network successfully learns to detect various semantic classes without direct supervision.
We consider the problem of domain adaptation in LiDAR-based 3D object detection. Towards this, we propose a simple yet effective training strategy called Gradual Batch Alternation that can adapt from a large labeled source domain to an insufficiently labeled target domain. The idea is to initiate the training with the batch of samples from the source and target domain data in an alternate fashion, but then gradually reduce the amount of the source domain data over time as the training progresses. This way the model slowly shifts towards the target domain and eventually better adapt to it. The domain adaptation experiments for 3D object detection on four benchmark autonomous driving datasets, namely ONCE, PandaSet, Waymo, and nuScenes, demonstrate significant performance gains over prior arts and strong baselines.
The problem of multi-object tracking is a fundamental computer vision research focus, widely used in public safety, transport, autonomous vehicles, robotics, and other regions involving artificial intelligence. Because of the complexity of natural scenes, object occlusion and semi-occlusion usually occur in fundamental tracking tasks. These can easily lead to ID switching, object loss, detect errors, and misaligned limitation boxes. These conditions have a significant impact on the precision of multi-object tracking. In this paper, we design a new multi-object tracker for the above issues that contains an object \textbf{Relative Location Mapping} (RLM) model and \textbf{Target Region Density} (TRD) model. The new tracker is more sensitive to the differences in position relationships between objects. It can introduce low-score detection frames into different regions in real-time according to the density of object regions in the video. This improves the accuracy of object tracking without consuming extensive arithmetic resources. Our study shows that the proposed model has considerably enhanced the HOTA and DF1 measurements on the MOT17 and MOT20 data sets when applied to the advanced MOT method.
Detecting occluded objects still remains a challenge for state-of-the-art object detectors. The objective of this work is to improve the detection for such objects, and thereby improve the overall performance of a modern object detector. To this end we make the following four contributions: (1) We propose a simple 'plugin' module for the detection head of two-stage object detectors to improve the recall of partially occluded objects. The module predicts a tri-layer of segmentation masks for the target object, the occluder and the occludee, and by doing so is able to better predict the mask of the target object. (2) We propose a scalable pipeline for generating training data for the module by using amodal completion of existing object detection and instance segmentation training datasets to establish occlusion relationships. (3) We also establish a COCO evaluation dataset to measure the recall performance of partially occluded and separated objects. (4) We show that the plugin module inserted into a two-stage detector can boost the performance significantly, by only fine-tuning the detection head, and with additional improvements if the entire architecture is fine-tuned. COCO results are reported for Mask R-CNN with Swin-T or Swin-S backbones, and Cascade Mask R-CNN with a Swin-B backbone.
This paper focuses on developing a more effective method of hierarchical propagation for semi-supervised Video Object Segmentation (VOS). Based on vision transformers, the recently-developed Associating Objects with Transformers (AOT) approach introduces hierarchical propagation into VOS and has shown promising results. The hierarchical propagation can gradually propagate information from past frames to the current frame and transfer the current frame feature from object-agnostic to object-specific. However, the increase of object-specific information will inevitably lead to the loss of object-agnostic visual information in deep propagation layers. To solve such a problem and further facilitate the learning of visual embeddings, this paper proposes a Decoupling Features in Hierarchical Propagation (DeAOT) approach. Firstly, DeAOT decouples the hierarchical propagation of object-agnostic and object-specific embeddings by handling them in two independent branches. Secondly, to compensate for the additional computation from dual-branch propagation, we propose an efficient module for constructing hierarchical propagation, i.e., Gated Propagation Module, which is carefully designed with single-head attention. Extensive experiments show that DeAOT significantly outperforms AOT in both accuracy and efficiency. On YouTube-VOS, DeAOT can achieve 86.0% at 22.4fps and 82.0% at 53.4fps. Without test-time augmentations, we achieve new state-of-the-art performance on four benchmarks, i.e., YouTube-VOS (86.2%), DAVIS 2017 (86.2%), DAVIS 2016 (92.9%), and VOT 2020 (0.622). Project page: https://github.com/z-x-yang/AOT.
Existing computer vision and object detection methods strongly rely on neural networks and deep learning. This active research area is used for applications such as autonomous driving, aerial photography, protection, and monitoring. Futuristic object detection methods rely on rectangular, boundary boxes drawn over an object to accurately locate its location. The modern object recognition algorithms, however, are vulnerable to multiple factors, such as illumination, occlusion, viewing angle, or camera rotation as well as cost. Therefore, deep learning-based object recognition will significantly increase the recognition speed and compatible external interference. In this study, we use convolutional neural networks (CNN) to recognize items, the neural networks have the advantages of end-to-end, sparse relation, and sharing weights. This article aims to classify the name of the various object based on the position of an object's detected box. Instead, under different distances, we can get recognition results with different confidence. Through this study, we find that this model's accuracy through recognition is mainly influenced by the proportion of objects and the number of samples. When we have a small proportion of an object on camera, then we get higher recognition accuracy; if we have a much small number of samples, we can get greater accuracy in recognition. The epidemic has a great impact on the world economy where designing a cheaper object recognition system is the need of time.
In this work, we propose a novel paradigm to encode the position of targets for target tracking in videos using transformers. The proposed paradigm, Dense Spatio-Temporal (DST) position encoding, encodes spatio-temporal position information in a pixel-wise dense fashion. The provided position encoding provides location information to associate targets across frames beyond appearance matching by comparing objects in two bounding boxes. Compared to the typical transformer positional encoding, our proposed encoding is applied to the 2D CNN features instead of the projected feature vectors to avoid losing positional information. Moreover, the designed DST encoding can represent the location of a single-frame object and the evolution of the location of the trajectory among frames uniformly. Integrated with the DST encoding, we build a transformer-based multi-object tracking model. The model takes a video clip as input and conducts the target association in the clip. It can also perform online inference by associating existing trajectories with objects from the new-coming frames. Experiments on video multi-object tracking (MOT) and multi-object tracking and segmentation (MOTS) datasets demonstrate the effectiveness of the proposed DST position encoding.
We present a differentiable pipeline for simulating the motion of objects that represent their geometry as a continuous density field parameterized as a deep network. This includes Neural Radiance Fields (NeRFs), and other related models. From the density field, we estimate the dynamical properties of the object, including its mass, center of mass, and inertia matrix. We then introduce a differentiable contact model based on the density field for computing normal and friction forces resulting from collisions. This allows a robot to autonomously build object models that are visually and \emph{dynamically} accurate from still images and videos of objects in motion. The resulting Dynamics-Augmented Neural Objects (DANOs) are simulated with an existing differentiable simulation engine, Dojo, interacting with other standard simulation objects, such as spheres, planes, and robots specified as URDFs. A robot can use this simulation to optimize grasps and manipulation trajectories of neural objects, or to improve the neural object models through gradient-based real-to-simulation transfer. We demonstrate the pipeline to learn the coefficient of friction of a bar of soap from a real video of the soap sliding on a table. We also learn the coefficient of friction and mass of a Stanford bunny through interactions with a Panda robot arm from synthetic data, and we optimize trajectories in simulation for the Panda arm to push the bunny to a goal location.
In this work we have determined the focal length of a concave lens by photographing the virtual image of an object by a smartphone camera. We have similarly determined the focal length of a convex lens by forming a virtual image of an object keeping it within the focal distance from the lens. When a photograph is taken by a smartphone, the transverse width of the image on the sensor of the camera in pixels can be read off by software available freely from the internet. By taking a photograph of the virtual image from two positions of the camera separated by a distance along the line of sight of the camera, we have determined the transverse width of the virtual image. From this we find the focal lengths of the lenses knowing the transverse width and the distance of the object from the lenses.
Multi-object tracking (MOT) is a vital component of intelligent video analytics applications such as surveillance and autonomous driving. The time and storage complexity required to execute deep learning models for visual object tracking hinder their adoption on embedded devices with limited computing power. In this paper, we aim to accelerate MOT by transferring the knowledge from high-level features of a complex network (teacher) to a lightweight network (student) at both training and inference times. The proposed AttTrack framework has three key components: 1) cross-model feature learning to align intermediate representations from the teacher and student models, 2) interleaving the execution of the two models at inference time, and 3) incorporating the updated predictions from the teacher model as prior knowledge to assist the student model. Experiments on pedestrian tracking tasks are conducted on the MOT17 and MOT15 datasets using two different object detection backbones YOLOv5 and DLA34 show that AttTrack can significantly improve student model tracking performance while sacrificing only minor degradation of tracking speed.
Deep neural networks are facing severe threats from adversarial attacks. Most existing black-box attacks fool target model by generating either global perturbations or local patches. However, both global perturbations and local patches easily cause annoying visual artifacts in adversarial example. Compared with some smooth regions of an image, the object region generally has more edges and a more complex texture. Thus small perturbations on it will be more imperceptible. On the other hand, the object region is undoubtfully the decisive part of an image to classification tasks. Motivated by these two facts, we propose an object-attentional adversarial attack method for untargeted attack. Specifically, we first generate an object region by intersecting the object detection region from YOLOv4 with the salient object detection (SOD) region from HVPNet. Furthermore, we design an activation strategy to avoid the reaction caused by the incomplete SOD. Then, we perform an adversarial attack only on the detected object region by leveraging Simple Black-box Adversarial Attack (SimBA). To verify the proposed method, we create a unique dataset by extracting all the images containing the object defined by COCO from ImageNet-1K, named COCO-Reduced-ImageNet in this paper. Experimental results on ImageNet-1K and COCO-Reduced-ImageNet show that under various system settings, our method yields the adversarial example with better perceptual quality meanwhile saving the query budget up to 24.16\% compared to the state-of-the-art approaches including SimBA.
Robust 3D object detection is critical for safe autonomous driving. Camera and radar sensors are synergistic as they capture complementary information and work well under different environmental conditions. Fusing camera and radar data is challenging, however, as each of the sensors lacks information along a perpendicular axis, that is, depth is unknown to camera and elevation is unknown to radar. We propose the camera-radar matching network CramNet, an efficient approach to fuse the sensor readings from camera and radar in a joint 3D space. To leverage radar range measurements for better camera depth predictions, we propose a novel ray-constrained cross-attention mechanism that resolves the ambiguity in the geometric correspondences between camera features and radar features. Our method supports training with sensor modality dropout, which leads to robust 3D object detection, even when a camera or radar sensor suddenly malfunctions on a vehicle. We demonstrate the effectiveness of our fusion approach through extensive experiments on the RADIATE dataset, one of the few large-scale datasets that provide radar radio frequency imagery. A camera-only variant of our method achieves competitive performance in monocular 3D object detection on the Waymo Open Dataset.
This paper presents the novel combination of a visual transformer style patch classifier with saccaded local attention. A novel optimisation paradigm for training object models is also presented, rather than the optimisation function minimising class membership probability error the network is trained to estimate the normalised distance to the centroid of labelled objects. This approach builds a degree of transnational invariance directly into the model and allows fast saccaded search with gradient ascent to find object centroids. The resulting saccaded visual transformer is demonstrated on human faces.
Knowledge distillation learns a lightweight student model that mimics a cumbersome teacher. Existing methods regard the knowledge as the feature of each instance or their relations, which is the instance-level knowledge only from the teacher model, i.e., the local knowledge. However, the empirical studies show that the local knowledge is much noisy in object detection tasks, especially on the blurred, occluded, or small instances. Thus, a more intrinsic approach is to measure the representations of instances w.r.t. a group of common basis vectors in the two feature spaces of the teacher and the student detectors, i.e., global knowledge. Then, the distilling algorithm can be applied as space alignment. To this end, a novel prototype generation module (PGM) is proposed to find the common basis vectors, dubbed prototypes, in the two feature spaces. Then, a robust distilling module (RDM) is applied to construct the global knowledge based on the prototypes and filtrate noisy global and local knowledge by measuring the discrepancy of the representations in two feature spaces. Experiments with Faster-RCNN and RetinaNet on PASCAL and COCO datasets show that our method achieves the best performance for distilling object detectors with various backbones, which even surpasses the performance of the teacher model. We also show that the existing methods can be easily combined with global knowledge and obtain further improvement. Code is available: https://github.com/hikvision-research/DAVAR-Lab-ML.
Image-sentence retrieval has attracted extensive research attention in multimedia and computer vision due to its promising application. The key issue lies in jointly learning the visual and textual representation to accurately estimate their similarity. To this end, the mainstream schema adopts an object-word based attention to calculate their relevance scores and refine their interactive representations with the attention features, which, however, neglects the context of the object representation on the inter-object relationship that matches the predicates in sentences. In this paper, we propose a Cross-modal Semantic Enhanced Interaction method, termed CMSEI for image-sentence retrieval, which correlates the intra- and inter-modal semantics between objects and words. In particular, we first design the intra-modal spatial and semantic graphs based reasoning to enhance the semantic representations of objects guided by the explicit relationships of the objects' spatial positions and their scene graph. Then the visual and textual semantic representations are refined jointly via the inter-modal interactive attention and the cross-modal alignment. To correlate the context of objects with the textual context, we further refine the visual semantic representation via the cross-level object-sentence and word-image based interactive attention. Experimental results on seven standard evaluation metrics show that the proposed CMSEI outperforms the state-of-the-art and the alternative approaches on MS-COCO and Flickr30K benchmarks.
Neural radiance fields (NeRFs) have recently emerged as a promising approach for 3D reconstruction and novel view synthesis. However, NeRF-based methods encode shape, reflectance, and illumination implicitly and this makes it challenging for users to manipulate these properties in the rendered images explicitly. Existing approaches only enable limited editing of the scene and deformation of the geometry. Furthermore, no existing work enables accurate scene illumination after object deformation. In this work, we introduce SPIDR, a new hybrid neural SDF representation. SPIDR combines point cloud and neural implicit representations to enable the reconstruction of higher quality object surfaces for geometry deformation and lighting estimation. meshes and surfaces for object deformation and lighting estimation. To more accurately capture environment illumination for scene relighting, we propose a novel neural implicit model to learn environment light. To enable more accurate illumination updates after deformation, we use the shadow mapping technique to approximate the light visibility updates caused by geometry editing. We demonstrate the effectiveness of SPIDR in enabling high quality geometry editing with more accurate updates to the illumination of the scene.
3D grasp synthesis generates grasping poses given an input object. Existing works tackle the problem by learning a direct mapping from objects to the distributions of grasping poses. However, because the physical contact is sensitive to small changes in pose, the high-nonlinear mapping between 3D object representation to valid poses is considerably non-smooth, leading to poor generation efficiency and restricted generality. To tackle the challenge, we introduce an intermediate variable for grasp contact areas to constrain the grasp generation; in other words, we factorize the mapping into two sequential stages by assuming that grasping poses are fully constrained given contact maps: 1) we first learn contact map distributions to generate the potential contact maps for grasps; 2) then learn a mapping from the contact maps to the grasping poses. Further, we propose a penetration-aware optimization with the generated contacts as a consistency constraint for grasp refinement. Extensive validations on two public datasets show that our method outperforms state-of-the-art methods regarding grasp generation on various metrics.
The ability to decompose complex natural scenes into meaningful object-centric abstractions lies at the core of human perception and reasoning. In the recent culmination of unsupervised object-centric learning, the Slot-Attention module has played an important role with its simple yet effective design and fostered many powerful variants. These methods, however, have been exceedingly difficult to train without supervision and are ambiguous in the notion of object, especially for complex natural scenes. In this paper, we propose to address these issues by investigating the potential of learnable queries as initializations for Slot-Attention learning, uniting it with efforts from existing attempts on improving Slot-Attention learning with bi-level optimization. With simple code adjustments on Slot-Attention, our model, Bi-level Optimized Query Slot Attention, achieves state-of-the-art results on 3 challenging synthetic and 7 complex real-world datasets in unsupervised image segmentation and reconstruction, outperforming previous baselines by a large margin. We provide thorough ablative studies to validate the necessity and effectiveness of our design. Additionally, our model exhibits great potential for concept binding and zero-shot learning. Our work is made publicly available at https://bo-qsa.github.io
Recent studies reveal that deep neural network (DNN) based object detectors are vulnerable to adversarial attacks in the form of adding the perturbation to the images, leading to the wrong output of object detectors. Most current existing works focus on generating perturbed images, also called adversarial examples, to fool object detectors. Though the generated adversarial examples themselves can remain a certain naturalness, most of them can still be easily observed by human eyes, which limits their further application in the real world. To alleviate this problem, we propose a differential evolution based dual adversarial camouflage (DE_DAC) method, composed of two stages to fool human eyes and object detectors simultaneously. Specifically, we try to obtain the camouflage texture, which can be rendered over the surface of the object. In the first stage, we optimize the global texture to minimize the discrepancy between the rendered object and the scene images, making human eyes difficult to distinguish. In the second stage, we design three loss functions to optimize the local texture, making object detectors ineffective. In addition, we introduce the differential evolution algorithm to search for the near-optimal areas of the object to attack, improving the adversarial performance under certain attack area limitations. Besides, we also study the performance of adaptive DE_DAC, which can be adapted to the environment. Experiments show that our proposed method could obtain a good trade-off between the fooling human eyes and object detectors under multiple specific scenes and objects.