Search

Unsupervised pretraining is of great significance for visual representation. Especially, contrastive learning has achieved great success recently, but existing approaches mostly ignored spatial information which is often crucial for visual representation. Strong semantic embedding has an inherent advantage for classification, but dense prediction tasks require more spatial and low-level representation. This paper presents heterogeneous contrastive learning (HCL), an effective approach that adds spatial information to the encoding stage to alleviate the learning inconsistency between the contrastive objective and strong data augmentation operations. We demonstrate the effectiveness of HCL by showing that (i) it achieves higher accuracy in instance discrimination, (ii) it surpasses existing pre-training methods in a series of downstream tasks (iii) and it shrinks the pre-training costs by half for almost 800 GPU-hours. More importantly, we show that our approach achieves higher efficiency in visual representations, and thus delivers a key message to inspire the future research of self-supervised visual representation learning.

Person re-identification (ReID) aims at matching identities across disjoint cameras. Its fundamental difficulty lies in associating images across individual cameras, where a key clue, i.e., identity appearance, is prone to the environmental factors of cameras and, consequently, subject to distinct image distributions due to the environmental differences between cameras. To associate images from training cameras, ReID methods strongly demand expensive inter-camera annotations for learning the relations between the distribution of these cameras, yet trained models are still not guaranteed to transfer well to unseen cameras. This problem significantly limits the application of ReID. This paper rethinks the working mechanism of conventional ReID approaches and puts forward a new solution. With an effective operator named Camera-based Batch Normalization (CBN), we guarantee an invariant input distribution independent of all cameras. Thus, the training and testing procedures are always conducted under the same input distribution. This alignment brings three benefits. First, ReID models enjoy better abilities to generalize across testing scenarios with unseen cameras and transfer across multiple training sets. Second, it makes better use of intra-camera annotations, which have been undervalued before due to the lack of cross-camera information. Ideally, the cost of inter-camera annotations can be largely reduced. Third, cross-modality tasks can be better defined through aligning visible/infrared cameras’ distributions. Experiments on a wide range of ReID tasks demonstrate the effectiveness of our approach.

Attention mechanisms have achieved success in video-based person re-identification (re-ID). However, current global attentions tend to focus on the most salient parts, e.g., clothes, and ignore other subtle but valuable cues, e.g., hair, bag, and shoes. They still do not make full use of valuable information from diverse parts of human bodies. To tackle this issue, we propose a Diverse Part Attentive Network (DPAN) to exploit discriminative and diverse body cues. The framework consists of two modules: spatial diverse part attention and temporal diverse part attention. The spatial module utilizes channel grouping to exploit diverse parts of human bodies including salient and subtle parts. The temporal module aims to learn diverse weights for fusing learned features. Besides, this framework is lightweight, which introduces marginal parameters and computational complexities. Extensive experiments were conducted on three popular benchmarks, i.e. iLIDS-VID, PRID2011 and MARS. Our method achieves competitive performance on these datasets compared with state-of-the-art methods.

N,N′-dimethyl-N-(4-pyridyl)-1,2-ethanediamine (DMPE) as a special N-alkylation precursor of 4-dimethylaminopyridine (DMAP) was grafted on the surface of γ-chloropropylated functionalized mesoporous silica MCM-41 to prepare immobilized DMAP. Characterization results of small angle X-ray diffraction, thermogravimetry and derivative thermogravimetry, Fourier transform infrared, scanning electron microscopy, energy-dispersive X-ray and transmission electron microscopy indicated that DMAP was successfully covalently bound to the surface of MCM-41. The highest loading of DMAP reached 0.9 mmol/g, which was 2.5 times higher than that of DMAP immobilized on the normal silica gel. The prepared immobilized DMAP as nucleophilic heterogeneous catalysts was successfully used for the acetylation reaction to produce α-tocopherol acetate. The highest yield reached 90.1% under the optimal condition which was summarized as follows: immobilized DMAP 100 mg, substrates mole ratio (acetic anhydride/α-tocopherol) 3:1, reaction temperature 50 ℃, stirring intensity 500 rpm, reaction in toluene for 12 h. The recyclability and stability of the DMAP grafted MCM-41 were excellent, as demonstrated by its use 20 times during batch operation without any loss of productivity.

Choline derivatives were covalently immobilized on the surface of γ -aminated silica. The obtained immobilized choline derivative was then successfully used for a transesterification reaction to produce glycerophosphocholine (GPC). Fourier transform infrared analysis and thermogravimetric analysis/differential thermal gravimetry indicated that the surface of the γ -aminated silica was covered by choline derivatives and the highest immobilization amount reached 1.1 mmol/g under optimal conditions. More importantly, the highest yield of GPC reached 97.9% during transesterification. With regard to GPC in food or medicine for human use, the immobilization technology can avoid catalyst contamination of the product and increase the safety of the product. The recyclability and stability of the immobilized choline derivative were excellent, as demonstrated by its use 20 times without any loss of productivity. A first-order kinetic model was employed and the relevant parameters were calculated to investigate kinetic characteristics of transesterification.

The huge variance of human pose and the misalign-ment of detected human images significantly increase the difficulty of pedestrian image matching in person Re-Identification (Re-ID). Moreover, the massive visual data being produced by surveillance video cameras requires highly efficient person Re-ID systems. Targeting to solve the first problem, this work proposes a robust and discriminative pedestrian image descriptor, namely, the Global–Local-Alignment Descriptor (GLAD). For the second problem, this work treats person Re-ID as image retrieval and proposes an efficient indexing and retrieval framework. GLAD explicitly leverages the local and global cues in the human body to generate a discriminative and robust representation. It consists of part extraction and descriptor learning modules, where several part regions are first detected and then deep neural networks are designed for representation learning on both the local and global regions. A hierarchical indexing and retrieval framework is designed to perform offline relevance mining to eliminate the huge person ID redundancy in the gallery set, and accelerate the online Re-ID procedure. Extensive experimental results on widely used public benchmark datasets show GLAD achieves competitive accuracy compared to the state-of-the-art methods. On a large-scale person, with the Re-ID dataset containing more than 520 K images, our retrieval framework significantly accelerates the online Re-ID procedure while also improving Re-ID accuracy. Therefore, this work has the potential to work better on person Re-ID tasks in real scenarios.

Catalytic efficiency in synthesis of vitamin E succinate was dramatically increased via the preparation of robust catalyst and the improvement of reaction system. 4-dimethylaminopyridine (DMAP) was covalently immobilized on nano-SiO2 to avoid the catalyst contamination of the product and permit the easy recycling of DMAP. Then, a hexane-acetone mixed solvent system was firstly introduced to replace the traditional single-solvent system, which was employed to improve the activity of immobilized DMAP derivative and the substrate solubility of the reaction system. The highest vitamin E succinate yield of 94% was achieved. In addition, the recyclability and stability of the immobilized DMAP derivative was excellent, the yield of vitamin E succinate had no obvious loss and remained 90% after recycling 20 times. The excellent results make this technology be a promising candidate for the industrial production of vitamin E succinate.

Automatic designing computationally efficient neural networks has received much attention in recent years. Existing approaches either utilize network pruning or leverage the network architecture search methods. This paper presents a new framework named network adjustment, which considers network accuracy as a function of FLOPs, so that under each network configuration, one can estimate the FLOPs utilization ratio (FUR) for each layer and use it to determine whether to increase or decrease the number of channels on the layer. Note that FUR, like the gradient of a non-linear function, is accurate only in a small neighborhood of the current network. Hence, we design an iterative mechanism so that the initial network undergoes a number of steps, each of which has a small 'adjusting rate' to control the changes to the network. The computational overhead of the entire search process is reasonable, i.e., comparable to that of re-training the final model from scratch. Experiments on standard image classification datasets and a wide range of base networks demonstrate the effectiveness of our approach, which consistently outperforms the pruning counterpart. The code is available at https://github.com/danczs/NetworkAdjustment.

AutoAugment has been a powerful algorithm that improves the accuracy of many vision tasks, yet it is sensitive to the operator space as well as hyper-parameters, and an improper setting may degenerate network optimization. This paper delves deep into the working mechanism, and reveals that AutoAugment may remove part of discriminative information from the training image and so insisting on the ground-truth label is no longer the best option. To relieve the inaccuracy of supervision, we make use of knowledge distillation that refers to the output of a teacher model to guide network training. Experiments are performed in standard image classification benchmarks, and demonstrate the effectiveness of our approach in suppressing noise of data augmentation and stabilizing training. Upon the cooperation of knowledge distillation and AutoAugment, we claim the new state-of-the-art on ImageNet classification with a top-1 accuracy of \(\mathbf {85.8\%}\).

The fundamental difficulty in person re-identification (ReID) lies in learning the correspondence among individual cameras. It strongly demands costly inter-camera annotations, yet the trained models are not guaranteed to transfer well to previously unseen cameras. These problems significantly limit the application of ReID. This paper rethinks the working mechanism of conventional ReID approaches and puts forward a new solution. With an effective operator named Camera-based Batch Normalization (CBN), we force the image data of all cameras to fall onto the same subspace, so that the distribution gap between any camera pair is largely shrunk. This alignment brings two benefits. First, the trained model enjoys better abilities to generalize across scenarios with unseen cameras as well as transfer across multiple training sets. Second, we can rely on intra-camera annotations, which have been undervalued before due to the lack of cross-camera information, to achieve competitive ReID performance. Experiments on a wide range of ReID tasks demonstrate the effectiveness of our approach. The code is available at https://github.com/automan000/Camera-based-Person-ReID.

Person re-identification (ReID) aims at finding the same person in different cameras. Training such systems usually requires a large amount of cross-camera pedestrians to be annotated from surveillance videos, which is labor-consuming especially when the number of cameras is large. Differently, this paper investigates ReID in an unexplored single-camera-training (SCT) setting, where each person in the training set appears in only one camera. To the best of our knowledge, this setting was never studied before. SCT enjoys the advantage of low-cost data collection and annotation, and thus eases ReID systems to be trained in a brand new environment. However, it raises major challenges due to the lack of cross-camera person occurrences, which conventional approaches heavily rely on to extract discriminative features. The key to dealing with the challenges in the SCT setting lies in designing an effective mechanism to complement cross-camera annotation. We start with a regular deep network for feature extraction, upon which we propose a novel loss function named multi-camera negative loss (MCNL). This is a metric learning loss motivated by probability, suggesting that in a multi-camera system, one image is more likely to be closer to the most similar negative sample in other cameras than to the most similar negative sample in the same camera. In experiments, MCNL significantly boosts ReID accuracy in the SCT setting, which paves the way of fast deployment of ReID systems with good performance on new target scenes.

With the capability of locating and tracking specific suspects or vehicles in a large camera network, person Re-Identification (ReID) and vehicle ReID show potential to be a key technology in smart surveillance system. They have been drawing lots of attentions from both academia and industry. To demonstrate our recent research progresses on those two tasks, we develop a robust and efficient person and video ReID system named as VP-ReID. This system is build based on our recent works including Deep Convolutional Neural Network design for discriminative feature extraction, efficient off-line indexing, as well as distance metric optimization for deep feature learning. Constructed upon those algorithms, VP-ReID identifies query vehicle and person efficiently and accurately from a large gallery set.

Although the performance of person Re-Identification (ReID) has been significantly boosted, many challenging issues in real scenarios have not been fully investigated, e.g., the complex scenes and lighting variations, viewpoint and pose changes, and the large number of identities in a camera network. To facilitate the research towards conquering those issues, this paper contributes a new dataset called MSMT17 with many important features, e.g., 1) the raw videos are taken by an 15-camera network deployed in both indoor and outdoor scenes, 2) the videos cover a long period of time and present complex lighting variations, and 3) it contains currently the largest number of annotated identities, i.e., 4,101 identities and 126,441 bounding boxes. We also observe that, domain gap commonly exists between datasets, which essentially causes severe performance drop when training and testing on different datasets. This results in that available training data cannot be effectively leveraged for new testing domains. To relieve the expensive costs of annotating new training samples, we propose a Person Transfer Generative Adversarial Network (PTGAN) to bridge the domain gap. Comprehensive experiments show that the domain gap could be substantially narrowed-down by the PTGAN., Comment: 10 pages, 9 figures; accepted in CVPR 2018

The huge variance of human pose and the misalignment of detected human images significantly increase the difficulty of person Re-Identification (Re-ID). Moreover, efficient Re-ID systems are required to cope with the massive visual data being produced by video surveillance systems. Targeting to solve these problems, this work proposes a Global-Local-Alignment Descriptor (GLAD) and an efficient indexing and retrieval framework, respectively. GLAD explicitly leverages the local and global cues in human body to generate a discriminative and robust representation. It consists of part extraction and descriptor learning modules, where several part regions are first detected and then deep neural networks are designed for representation learning on both the local and global regions. A hierarchical indexing and retrieval framework is designed to eliminate the huge redundancy in the gallery set, and accelerate the online Re-ID procedure. Extensive experimental results show GLAD achieves competitive accuracy compared to the state-of-the-art methods. Our retrieval framework significantly accelerates the online Re-ID procedure without loss of accuracy. Therefore, this work has potential to work better on person Re-ID tasks in real scenarios., Comment: Accepted by ACM MM2017, 9 pages, 5 figures