No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
ISAR: A Benchmark for Single- and Few-Shot Object Instance Segmentation and Re-Identification
... The combination of class-agnostic detection and some kind of object re-identification has also been explored in related work. A two-step approach in which objects are first class-agnostically localized based on the Segment Anything Model (SAM) [23] and then feature vectors of DINO v2 [57] are used for re-identifying the novel categories is described in [58]. However, it should be noted that all experiments are performed on synthesized data, which limits the value of the results. ...
... The re-identification of objects in industrial and assembly scenarios after class-agnostic detection has been evaluated in [58,62]. Gorlo et al. [58] train and test only on simulated scenes. ...
... The re-identification of objects in industrial and assembly scenarios after class-agnostic detection has been evaluated in [58,62]. Gorlo et al. [58] train and test only on simulated scenes. Tests on real-world images were not part of their experiments, which limits the meaningfulness of the results. ...
Object detection is a crucial capability of autonomous agents for human–robot collaboration, as it facilitates the identification of the current processing state. In industrial scenarios, it is uncommon to have comprehensive knowledge of all the objects involved in a given task. Furthermore, training during deployment is not a viable option. Consequently, there is a need for a detector that is able to adapt to novel objects during deployment without the necessity of retraining or fine-tuning on novel data. To achieve this, we propose to exploit the ability of discriminative embeddings learned by an object re-identification model to generalize to unknown categories described by a few shots. To do so, we extract object crops with a class-agnostic detector and then compare the object features with the prototypes of the novel objects. Moreover, we demonstrate that the embedding is also effective for predicting regions of interest, which narrows the search space of the class-agnostic detector and, consequently, increases processing speed. The effectiveness of our approach is evaluated in an assembly scenario, wherein the majority of objects belong to categories distinct from those present in the training datasets. Our experiments demonstrate that, in this scenario, our approach outperforms the current best few-shot object-detection approach DE-ViT, which also does not perform fine-tuning on novel data, in terms of both detection capability and inference speed.
Can our video understanding systems perceive objects when a heavy occlusion exists in a scene? To answer this question, we collect a large-scale dataset called OVIS for occluded video instance segmentation, that is, to simultaneously detect, segment, and track instances in occluded scenes. OVIS consists of 296k high-quality instance masks from 25 semantic categories, where object occlusions usually occur. While our human vision systems can understand those occluded instances by contextual reasoning and association, our experiments suggest that current video understanding systems cannot. On the OVIS dataset, the highest AP achieved by state-of-the-art algorithms is only 16.3, which reveals that we are still at a nascent stage for understanding objects, instances, and videos in a real-world scenario. We also present a simple plug-and-play module that performs temporal feature calibration to complement missing object cues caused by occlusion. Built upon MaskTrack R-CNN and SipMask, we obtain a remarkable AP improvement on the OVIS dataset. The OVIS dataset and project code are available at http://songbai.site/ovis .
Semantic segmentation assigns a class label to each image pixel. This dense prediction problem requires large amounts of manually annotated data, which is often unavailable. Few-shot learning aims to learn the pattern of a new category with only a few annotated examples. In this paper, we formulate the few-shot semantic segmentation problem from 1-way (class) to N-way (classes). Inspired by few-shot classification, we propose a generalized framework for few-shot semantic segmentation with an alternative training scheme. The framework is based on prototype learning and metric learning. Our approach outperforms the baselines by a large margin and shows comparable performance for 1-way few-shot semantic segmentation on PASCAL VOC 2012 dataset.
The Visual Object Tracking challenge VOT2020 is the eighth annual tracker benchmarking activity organized by the VOT initiative. Results of 58 trackers are presented; many are state-of-the-art trackers published at major computer vision conferences or in journals in the recent years. The VOT2020 challenge was composed of five sub-challenges focusing on different tracking domains: (i) VOT-ST2020 challenge focused on short-term tracking in RGB, (ii) VOT-RT2020 challenge focused on “real-time” short-term tracking in RGB, (iii) VOT-LT2020 focused on long-term tracking namely coping with target disappearance and reappearance, (iv) VOT-RGBT2020 challenge focused on short-term tracking in RGB and thermal imagery and (v) VOT-RGBD2020 challenge focused on long-term tracking in RGB and depth imagery. Only the VOT-ST2020 datasets were refreshed. A significant novelty is introduction of a new VOT short-term tracking evaluation methodology, and introduction of segmentation ground truth in the VOT-ST2020 challenge – bounding boxes will no longer be used in the VOT-ST challenges. A new VOT Python toolkit that implements all these novelites was introduced. Performance of the tested trackers typically by far exceeds standard baselines. The source code for most of the trackers is publicly available from the VOT page. The dataset, the evaluation kit and the results are publicly available at the challenge website (http://votchallenge.net).
Significant progress has been made recently in developing few-shot object segmentation methods. Learning is shown to be successful in few-shot segmentation settings, using pixel-level, scribbles and bounding box supervision. This paper takes another approach, i.e., only requiring image-level label for few-shot object segmentation. We propose a novel multi-modal interaction module for few-shot object segmentation that utilizes a co-attention mechanism using both visual and word embedding. Our model using image-level labels achieves 4.8% improvement over previously proposed image-level few-shot object segmentation. It also outperforms state-of-the-art methods that use weak bounding box supervision on PASCAL-5^i. Our results show that few-shot segmentation benefits from utilizing word embeddings, and that we are able to perform few-shot segmentation using stacked joint visual semantic processing with weak image-level labels. We further propose a novel setup, Temporal Object Segmentation for Few-shot Learning (TOSFL) for videos. TOSFL can be used on a variety of public video data such as Youtube-VOS, as demonstrated in both instance-level and category-level TOSFL experiments.
Although great progress in supervised person re-identification (Re-ID) has been made recently, due to the viewpoint variation of a person, Re-ID remains a massive visual challenge. Most existing viewpoint-based person Re-ID methods project images from each viewpoint into separated and unrelated sub-feature spaces. They only model the identity-level distribution inside an individual viewpoint but ignore the underlying relationship between different viewpoints. To address this problem, we propose a novel approach, called Viewpoint-Aware Loss with Angular Regularization (VA-reID). Instead of one subspace for each viewpoint, our method projects the feature from different viewpoints into a unified hypersphere and effectively models the feature distribution on both the identity-level and the viewpoint-level. In addition, rather than modeling different viewpoints as hard labels used for conventional viewpoint classification, we introduce viewpoint-aware adaptive label smoothing regularization (VALSR) that assigns the adaptive soft label to feature representation. VALSR can effectively solve the ambiguity of the viewpoint cluster label assignment. Extensive experiments on the Market1501 and DukeMTMC-reID datasets demonstrated that our method outperforms the state-of-the-art supervised Re-ID methods.
Simultaneous Localization and Mapping (SLAM) plays an important role in the computer vision and robotics field. The traditional SLAM framework adopts a strong static world assumption for analysis convenience. How to cope with dynamic environments is of vital importance and attracts more attentions. Existing SLAM systems toward dynamic scenes either solely utilize semantic information, solely utilize geometry information, or naively combine the results from them in a loosely coupled way. In this paper, we present SOF-SLAM: Semantic Optical Flow SLAM, a visual semantic SLAM system toward dynamic environments, which is built on RGB-D mode of ORB-SLAM2. A new dynamic features detection approach called semantic optical flow is proposed, which is a kind of tightly coupled way and can fully take advantage of feature’s dynamic characteristic hidden in semantic and geometry information to remove dynamic features effectively and reasonably. The pixel-wise semantic segmentation results generated by SegNet serve as mask in the proposed semantic optical flow to get a reliable fundamental matrix, which is then used to filter out the truly dynamic features. Only the remaining static features are reserved in the tracking and optimization module to achieve accurate camera pose estimation in dynamic environments. Experiments on public TUM RGB-D dataset and in real-world environment are conducted. Compared with ORB-SLAM2, the proposed SOF-SLAM achieves averagely 96.73% improvements in high-dynamic scenarios. It also outperforms the other four state-of-the-art SLAM systems which cope with the dynamic environments.
The last decade has witnessed a growing interest in video salient object detection (VSOD). However, the research community long-term lacked a well-established VSOD dataset representative of real dynamic scenes with high-quality annotations. To address this issue, we elaborately collected a visual-attention-consistent Densely Annotated VSOD (DAVSOD) dataset, which contains 226 videos with 23,938 frames that cover diverse realistic-scenes, objects , instances and motions. With corresponding real human eye-fixation data, we obtain precise ground-truths. This is the first work that explicitly emphasizes the challenge of saliency shift, i.e., the video salient object(s) may dynamically change. To further contribute the community a complete benchmark, we systematically assess 17 representative VSOD algorithms over seven existing VSOD datasets and our DAVSOD with totally ∼84K frames (largest-scale). Utilizing three famous metrics, we then present a comprehensive and insightful performance analysis. Furthermore, we propose a baseline model. It is equipped with a saliency-shift-aware convLSTM, which can efficiently capture video saliency dynamics through learning human attention-shift behavior. Extensive experiments 1 open up promising future directions for model development and comparison.
Current few-shot segmentation (FSS) approaches have made tremendous achievements based on prototypical learning techniques. However, due to the scarcity of the support data provided, FSS methods still suffer from the intra-class and inter-class gaps. In this paper, we propose a uniform network to fill both the gaps, termed FGNet. It consists of the novel design of a Self-Adaptive Module (SAM) to emphasize the query feature to generate an enhanced prototype for self-alignment. Such a prototype caters to each query sample itself since it contains the underlying intra-instance information, which gets around the intra-class appearance gap. Moreover, we design an Inter-class Feature Separation Module (IFSM) to separate the feature space of the target class from other classes, which contributes to bridging the inter-class gap. In addition, we present several new losses and a method termed B-SLIC, which help to further enhance the separation performance of FGNet. Experimental results show that FGNet reduces both the gaps for FSS by SAM and IFSM respectively, and achieves state-of-the-art performances on both PASCAL-5i and COCO-20i datasets compared with previous top-performing approaches.
Existing few-shot segmentation methods have achieved great progress based on the support-query matching framework. But they still heavily suffer from the limited coverage of intra-class variations from the few-shot supports provided. Motivated by the simple Gestalt principle that pixels belonging to the same object are more similar than those to different objects of same class, we propose a novel self-support matching strategy to alleviate this problem, which uses query prototypes to match query features, where the query prototypes are collected from high-confidence query predictions. This strategy can effectively capture the consistent underlying characteristics of the query objects, and thus fittingly match query features. We also propose an adaptive self-support background prototype generation module and self-support loss to further facilitate the self-support matching procedure. Our self-support network substantially improves the prototype quality, benefits more improvement from stronger backbones and more supports, and achieves SOTA on multiple datasets. Codes are at https://github.com/fanq15/SSP.KeywordsFew-shot semantic segmentationSelf-support prototype (SSP)Self-support matchingAdaptive background prototype generation
Current object detectors are limited in vocabulary size due to the small scale of detection datasets. Image classifiers, on the other hand, reason about much larger vocabularies, as their datasets are larger and easier to collect. We propose Detic, which simply trains the classifiers of a detector on image classification data and thus expands the vocabulary of detectors to tens of thousands of concepts. Unlike prior work, Detic does not need complex assignment schemes to assign image labels to boxes based on model predictions, making it much easier to implement and compatible with a range of detection architectures and backbones. Our results show that Detic yields excellent detectors even for classes without box annotations. It outperforms prior work on both open-vocabulary and long-tail detection benchmarks. Detic provides a gain of 2.4 mAP for all classes and 8.3 mAP for novel classes on the open-vocabulary LVIS benchmark. On the standard LVIS benchmark, Detic obtains 41.7 mAP when evaluated on all classes, or only rare classes, hence closing the gap in performance for object categories with few samples. For the first time, we train a detector with all the twenty-one-thousand classes of the ImageNet dataset and show that it generalizes to new datasets without finetuning. Code is available at https://github.com/facebookresearch/Detic.
Recently, several spatial-temporal memory-based methods have verified that storing intermediate frames and their masks as memory are helpful to segment target objects in videos. However, they mainly focus on better matching between the current frame and the memory frames without explicitly paying attention to the quality of the memory. Therefore, frames with poor segmentation masks are prone to be memorized, which leads to a segmentation mask error accumulation problem and further affect the segmentation performance. In addition, the linear increase of memory frames with the growth of frame number also limits the ability of the models to handle long videos. To this end, we propose a Quality-aware Dynamic Memory Network (QDMN) to evaluate the segmentation quality of each frame, allowing the memory bank to selectively store accurately segmented frames to prevent the error accumulation problem. Then, we combine the segmentation quality with temporal consistency to dynamically update the memory bank to improve the practicability of the models. Without any bells and whistles, our QDMN achieves new state-of-the-art performance on both DAVIS and YouTube-VOS benchmarks. Moreover, extensive experiments demonstrate that the proposed Quality Assessment Module (QAM) can be applied to memory-based methods as generic plugins and significantly improves performance. Our source code is available at https://github.com/workforai/QDMN.KeywordsVideo object segmentationMemory bank
We present XMem, a video object segmentation architecture for long videos with unified feature memory stores inspired by the Atkinson-Shiffrin memory model. Prior work on video object segmentation typically only uses one type of feature memory. For videos longer than a minute, a single feature memory model tightly links memory consumption and accuracy. In contrast, following the Atkinson-Shiffrin model, we develop an architecture that incorporates multiple independent yet deeply-connected feature memory stores: a rapidly updated sensory memory, a high-resolution working memory, and a compact thus sustained long-term memory. Crucially, we develop a memory potentiation algorithm that routinely consolidates actively used working memory elements into the long-term memory, which avoids memory explosion and minimizes performance decay for long-term prediction. Combined with a new memory reading mechanism, XMem greatly exceeds state-of-the-art performance on long-video datasets while being on par with state-of-the-art methods (that do not work on long videos) on short-video datasets.
We present YOFO (You Only inFer Once), a new paradigm for referring video object segmentation (RVOS) that operates in an one-stage manner. Our key insight is that the language descriptor should serve as target-specific guidance to identify the target object, while a direct feature fusion of image and language can increase feature complexity and thus may be sub-optimal for RVOS. To this end, we propose a meta-transfer module, which is trained in a learning-to-learn fashion and aims to transfer the target-specific information from the language domain to the image domain, while discarding the uncorrelated complex variations of language description. To bridge the gap between the image and language domains, we develop a multi-scale cross-modal feature mining block that aggregates all the essential features required by RVOS from both domains and generates regression labels for the meta-transfer module. The whole system can be trained in an end-to-end manner and shows competitive performance against state-of-the-art two-stage approaches.
Image retrieval task consists of finding similar images to a query image from a set of gallery (database) images. Such systems are used in various applications e.g. person re-identification (ReID) or visual product search. Despite active development of retrieval models it still remains a challenging task mainly due to large intra-class variance caused by changes in view angle, lighting, background clutter or occlusion, while inter-class variance may be relatively low. A large portion of current research focuses on creating more robust features and modifying objective functions, usually based on Triplet Loss. Some works experiment with using centroid/proxy representation of a class to alleviate problems with computing speed and hard samples mining used with Triplet Loss. However, these approaches are used for training alone and discarded during the retrieval stage. In this paper we propose to use the mean centroid representation both during training and retrieval. Such an aggregated representation is more robust to outliers and assures more stable features. As each class is represented by a single embedding - the class centroid - both retrieval time and storage requirements are reduced significantly. Aggregating multiple embeddings results in a significant reduction of the search space due to lowering the number of candidate target vectors, which makes the method especially suitable for production deployments. Comprehensive experiments conducted on two ReID and Fashion Retrieval datasets demonstrate effectiveness of our method, which outperforms the current state-of-the-art. We propose centroid training and retrieval as a viable method for both Fashion Retrieval and ReID applications. Our code is available at https://github.com/mikwieczorek/centroids-reid.
Vehicle Re-Identification (V-ReID) is a critical task that associates the same vehicle across images from different camera viewpoints. Many works explore attribute clues to enhance V-ReID; however, there is usually a lack of effective interaction between the attribute-related modules and final V-ReID objective. In this work, we propose a new method to efficiently explore discriminative information from vehicle attributes (for instance, color and type). We introduce AttributeNet (ANet) that jointly extracts identity-relevant features and attribute features. We enable the interaction by distilling the ReID-helpful attribute feature and adding it into the general ReID feature to increase the discrimination power. Moreover, we propose a constraint, named Amelioration Constraint (AC), which encourages the feature after adding attribute features onto the general ReID feature to be more discriminative than the original general ReID feature. We validate the effectiveness of our framework on three challenging datasets. Experimental results show that our method achieves the state-of-the-art performance.
The assumption of scene rigidity is common in visual SLAM algorithms. However, it limits their applicability in populated real-world environments. Furthermore, most scenarios including autonomous driving, multi-robot collaboration and augmented/virtual reality, require explicit motion information of the surroundings to help with decision making and scene understanding. We present in this paper DynaSLAM II, a visual SLAM system for stereo and RGB-D configurations that tightly integrates the multi-object tracking capability. DynaSLAM II makes use of instance semantic segmentation and of ORB features to track dynamic objects. The structure of the static scene and of the dynamic objects is optimized jointly with the trajectories of both the camera and the moving agents within a novel bundle adjustment proposal. The 3D bounding boxes of the objects are also estimated and loosely optimized within a fixed temporal window. We demonstrate that tracking dynamic objects does not only provide rich clues for scene understanding but is also beneficial for camera tracking. The project code will be released upon acceptance.
Manually labeling video datasets for segmentation tasks is extremely time consuming. We introduce ScribbleBox, an interactive framework for annotating object instances with masks in videos with a significant boost in efficiency. In particular, we split annotation into two steps: annotating objects with tracked boxes, and labeling masks inside these tracks. We introduce automation and interaction in both steps. Box tracks are annotated efficiently by approximating the trajectory using a parametric curve with a small number of control points which the annotator can interactively correct. Our approach tolerates a modest amount of noise in box placements, thus typically requiring only a few clicks to annotate a track to a sufficient accuracy. Segmentation masks are corrected via scribbles which are propagated through time. We show significant performance gains in annotation efficiency over past work. We show that our ScribbleBox approach reaches 88.92% J&F on DAVIS2017 with an average of 9.14 clicks per box track, and only 4 frames requiring scribble annotation in a video of 65.3 frames on average.
Semi-supervised video object segmentation (VOS) is a task that involves predicting a target object in a video when the ground truth segmentation mask of the target object is given in the first frame. Recently, space-time memory networks (STM) have received significant attention as a promising solution for semi-supervised VOS. However, an important point is overlooked when applying STM to VOS. The solution (STM) is non-local, but the problem (VOS) is predominantly local. To solve the mismatch between STM and VOS, we propose a kernelized memory network (KMN). Before being trained on real videos, our KMN is pre-trained on static images, as in previous works. Unlike in previous works, we use the Hide-and-Seek strategy in pre-training to obtain the best possible results in handling occlusions and segment boundary extraction. The proposed KMN surpasses the state-of-the-art on standard benchmarks by a significant margin (+5% on DAVIS 2017 test-dev set). In addition, the runtime of KMN is 0.12 s per frame on the DAVIS 2016 validation set, and the KMN rarely requires extra computation, when compared with STM.
We propose a unified referring video object segmentation network (URVOS). URVOS takes a video and a referring expression as inputs, and estimates the object masks referred by the given language expression in the whole video frames. Our algorithm addresses the challenging problem by performing language-based object segmentation and mask propagation jointly using a single deep neural network with a proper combination of two attention models. In addition, we construct the first large-scale referring video object segmentation dataset called Refer-Youtube-VOS. We evaluate our model on two benchmark datasets including ours and demonstrate the effectiveness of the proposed approach. The dataset is released at https://github.com/skynbe/Refer-Youtube-VOS.
We present a new method that views object detection as a direct set prediction problem. Our approach streamlines the detection pipeline, effectively removing the need for many hand-designed components like a non-maximum suppression procedure or anchor generation that explicitly encode our prior knowledge about the task. The main ingredients of the new framework, called DEtection TRansformer or DETR, are a set-based global loss that forces unique predictions via bipartite matching, and a transformer encoder-decoder architecture. Given a fixed small set of learned object queries, DETR reasons about the relations of the objects and the global image context to directly output the final set of predictions in parallel. The new model is conceptually simple and does not require a specialized library, unlike many other modern detectors. DETR demonstrates accuracy and run-time performance on par with the well-established and highly-optimized Faster R-CNN baseline on the challenging COCO object detection dataset. Moreover, DETR can be easily generalized to produce panoptic segmentation in a unified manner. We show that it significantly outperforms competitive baselines. Training code and pretrained models are available at https://github.com/facebookresearch/detr.
One fundamental challenge of vehicle re-identification (re-id) is to learn robust and discriminative visual representation, given the significant intra-class vehicle variations across different camera views. As the existing vehicle datasets are limited in terms of training images and viewpoints, we propose to build a unique large-scale vehicle dataset (called VehicleNet) by harnessing four public vehicle datasets, and design a simple yet effective two-stage progressive approach to learning more robust visual representation from VehicleNet. The first stage of our approach is to learn the generic representation for all domains (i.e., source vehicle datasets) by training with the conventional classification loss. This stage relaxes the full alignment between the training and testing domains, as it is agnostic to the target vehicle domain. The second stage is to fine-tune the trained model purely based on the target vehicle set, by minimizing the distribution discrepancy between our VehicleNet and any target domain. We discuss our proposed multi-source dataset VehicleNet and evaluate the effectiveness of the two-stage progressive representation learning through extensive experiments. We achieve the state-of-art accuracy of
mAP on the private test set of AICity Challenge, and competitive results on two other public vehicle re-id datasets, i.e., VeRi-776 and VehicleID. We hope this new VehicleNet dataset and the learned robust representations can pave the way for vehicle re-id in the real-world environments.
State-of-the-art semantic segmentation methods require sufficient labeled data to achieve good results and hardly work on unseen classes without fine-tuning. Few-shot segmentation is thus proposed to tackle this problem by learning a model that quickly adapts to new classes with a few labeled support samples. Theses frameworks still face the challenge of generalization ability reduction on unseen classes due to inappropriate use of high-level semantic information of training classes and spatial inconsistency between query and support targets. To alleviate these issues, we propose the Prior Guided Feature Enrichment Network (PFENet). It consists of novel designs of (1) a training-free prior mask generation method that not only retains generalization power but also improves model performance and (2) Feature Enrichment Module (FEM) that overcomes spatial inconsistency by adaptively enriching query features with support features and prior masks. Extensive experiments on PASCAL-
and COCO prove that the proposed prior generation method and FEM both improve the baseline method significantly. Our PFENet also outperforms state-of-the-art methods by a large margin without efficiency loss. It is surprising that our model even generalizes to cases without labeled support samples.
Few-shot segmentation aims at assigning a category label to each image pixel with few annotated samples. It is a challenging task since the dense prediction can only be achieved under the guidance of latent features defined by sparse annotations. Existing meta-learning based method tends to fail in generating category-specifically discriminative descriptor when the visual features extracted from support images are marginalized in embedding space. To address this issue, this paper presents an adaptive tuning framework, in which the distribution of latent features across different episodes is dynamically adjusted based on a self-segmentation scheme, augmenting category-specific descriptors for label prediction. Specifically, a novel self-supervised inner-loop is firstly devised as the base learner to extract the underlying semantic features from the support image. Then, gradient maps are calculated by back-propagating self-supervised loss through the obtained features, and leveraged as guidance for augmenting the corresponding elements in the embedding space. Finally, with the ability to continuously learn from different episodes, an optimization-based meta-learner is adopted as outer loop of our proposed framework to gradually refine the segmentation results. Extensive experiments on benchmark PASCAL-5i and COCO-20i datasets demonstrate the superiority of our proposed method over state-of-the-art.
Most of current person re-identification (ReID) methods neglect a spatial-temporal constraint. Given a query image, conventional methods compute the feature distances between the query image and all the gallery images and return a similarity ranked table. When the gallery database is very large in practice, these approaches fail to obtain a good performance due to appearance ambiguity across different camera views. In this paper, we propose a novel two-stream spatial-temporal person ReID (st-ReID) framework that mines both visual semantic information and spatial-temporal information. To this end, a joint similarity metric with Logistic Smoothing (LS) is introduced to integrate two kinds of heterogeneous information into a unified framework. To approximate a complex spatial-temporal probability distribution, we develop a fast Histogram-Parzen (HP) method. With the help of the spatial-temporal constraint, the st-ReID model eliminates lots of irrelevant images and thus narrows the gallery database. Without bells and whistles, our st-ReID method achieves rank-1 accuracy of 98.1% on Market-1501 and 94.4% on DukeMTMC-reID, improving from the baselines 91.2% and 83.8%, respectively, outperforming all previous state-of-theart methods by a large margin.
