The giant panda (Ailuropoda melanoleuca) is one of the animals with the largest body weight differences between its birth and adult stages, where the newborn cub is 0.1% the size of its mother. The rapid growth of panda cubs has been reported previously, but little is known about the growth pattern of their entire lifetime. In this study, we analyzed body weight records from 206 captive giant pandas collected between 2000 and 2022. The dataset comprised 26,340 measurements, including 12,314 records from 98 males and 14,026 from 108 females, with the oldest male and female being 32 and 37 years old, respectively. Utilizing machine learning, we predicted daily body weights throughout the pandas’ lifespan, allowing us to establish detailed daily normative body weight ranges from birth to advanced age. This study presents the first comprehensive analysis of lifetime body weight distribution in giant pandas, enhancing our understanding of their developmental biology and informing improved body weight management strategies for captive populations.

We observe an elusive defect in face detectors which is overlooked in existing works. Specifically, we find that face detectors output unstable confidence scores when faces are slightly shifted in position. The confidence scores of the shifted faces can be lower than the detection threshold and result in false negatives. We define this phenomenon as face detection spatial instability. In essence, detectors face the spatial instability problem because they perform badly for challenging object positions. An object position is challenging when it is not sufficiently modeled by any existing anchors. To deal with this problem, we propose Matching Similarity Auxiliary (MSA) box, which consists of three parts: MSA assign, Image Digitization Compensator (IDC), and Soft Smooth L1 Loss (SSL-Loss). Specifically, MSA assign and IDC are designed to dig more challenging samples, which guide the face detectors to enhance detection performance in these extreme cases. SSL-Loss balances the training phase based on the regression efforts of each target. It improves the localization performance even with more challenging positive samples involved. Experiments demonstrate that our MSAbox achieves spatially stable face detection.

This paper proposes a novel Unified Feature Optimization (UFO) paradigm for training and deploying deep models under real-world and large-scale scenarios, which requires a collection of multiple AI functions. UFO aims to benefit each single task with a large-scale pretraining on all tasks. Compared with existing foundation models, UFO has two points of emphasis, i.e., relatively smaller model size and NO adaptation cost: 1) UFO squeezes a wide range of tasks into a moderate-sized unified model in a multi-task learning manner and further trims the model size when transferred to down-stream tasks. 2) UFO does not emphasize transfer to novel tasks. Instead, it aims to make the trimmed model dedicated for one or more already-seen task. To this end, it directly selects partial modules in the unified model, requiring completely NO adaptation cost. With these two characteristics, UFO provides great convenience for flexible deployment, while maintaining the benefits of large-scale pretraining. A key merit of UFO is that the trimming process not only reduces the model size and inference consumption, but also even improves the accuracy on certain tasks. Specifically, UFO considers the multi-task training and brings a two-fold impact on the unified model: some closely-related tasks have mutual benefits, while some tasks have conflicts against each other. UFO manages to reduce the conflicts and preserve the mutual benefits through a novel Network Architecture Search (NAS) method. Experiments on a wide range of deep representation learning tasks (i.e., face recognition, person re-identification, vehicle re-identification and product retrieval) show that the model trimmed from UFO achieves higher accuracy than its single-task-trained counterpart and yet has smaller model size, validating the concept of UFO. Besides, UFO also supported the release of 17 billion parameters computer vision (CV) foundation model which is the largest CV model in the industry. Code: https://github.com/PaddlePaddle/VIMER/tree/main/UFO .

Learning discriminative representation using large-scale face datasets in the wild is crucial for real-world applications, yet it remains challenging. The difficulties lie in many aspects and this work focus on computing resource constraint and long-tailed class distribution. Recently, classification-based representation learning with deep neural networks and well-designed losses have demonstrated good recognition performance. However, the computing and memory cost linearly scales up to the number of identities (classes) in the training set, and the learning process suffers from unbalanced classes. In this work, we propose a dynamic class queue (DCQ) to tackle these two problems. Specifically, for each iteration during training, a subset of classes for recognition are dynamically selected and their class weights are dynamically generated on-the-fly which are stored in a queue. Since only a subset of classes is selected for each iteration, the computing requirement is reduced. By using a single server without model parallel, we empirically verify in large-scale datasets that 10% of classes are sufficient to achieve similar performance as using all classes. Moreover, the class weights are dynamically generated in a few-shot manner and therefore suitable for tail classes with only a few instances. We show clear improvement over a strong baseline in the largest public dataset Megaface Challenge2 (MF2) which has 672K identities and over 88% of them have less than 10 instances. Code is available at https://github.com/bilylee/DCQ

Fast appearance variations and the distractions of similar objects are two of the most challenging problems in visual object tracking. Unlike many existing trackers that focus on modeling only the target, in this work, we consider the transient variations of the whole scene. The key insight is that the object correspondence and spatial layout of the whole scene are consistent (i.e., global structure consistency) in consecutive frames which helps to disambiguate the target from distractors. Moreover, modeling transient variations enables to localize the target under fast variations. Specifically, we propose an effective and efficient short-term model that learns to exploit the global structure consistency in a short time and thus can handle fast variations and distractors. Since short-term modeling falls short of handling occlusion and out of the views, we adopt the long-short term paradigm and use a long-term model that corrects the short-term model when it drifts away from the target or the target is not present. These two components are carefully combined to achieve the balance of stability and plasticity during tracking. We empirically verify that the proposed tracker can tackle the two challenging scenarios and validate it on large scale benchmarks. Remarkably, our tracker improves state-of-the-art-performance on VOT2018 from 0.440 to 0.460, GOT-10k from 0.611 to 0.640, and NFS from 0.619 to 0.629.

Model update lies at the heart of object tracking. Generally, model update is formulated as an online learning problem where a target model is learned over the online training set. Our key innovation is to formulate the model update problem in the meta-learning framework and learn the online learning algorithm itself using large numbers of offline videos, i.e., learning to update. The learned updater takes as input the online training set and outputs an updated target model. As a first attempt, we design the learned updater based on recurrent neural networks (RNNs) and demonstrate its application in a template-based tracker and a correlation filter-based tracker. Our learned updater consistently improves the base trackers and runs faster than realtime on GPU while requiring small memory footprint during testing. Experiments on standard benchmarks demonstrate that our learned updater outperforms commonly used update baselines including the efficient exponential moving average (EMA)-based update and the well-designed stochastic gradient descent (SGD)-based update. Equipped with our learned updater, the template-based tracker achieves state-of-the-art performance among realtime trackers on GPU.

A facial landmark localization method is introduced in this paper. It consists of several models that developed on the basis of stacked hourglass network and performs landmark localization from coarse to fine in two stages. The results are fused according to expectation consensus to remove outliers and it can alleviate the problem of inductive bias. We improve the alignment model with weighted heatmaps to get rid of the quantization errors and by applying auto ML to search better network structures under accurate guidance. Besides, the models are trained with face segmentation and a well-designed augmentation scheme. The method achieves the 1st place in the Grand Challenge of 106-point Facial Landmark Localization in ICME2019 with AUC of 84.01%.

Facial landmark localization is a very crucial step in numerous face related applications, such as face recognition, facial pose estimation, face image synthesis, etc. However, previous competitions on facial landmark localization (i.e., the 300-W, 300-VW and Menpo challenges) aim to predict 68-point landmarks, which are incompetent to depict the structure of facial components. In order to overcome this problem, we construct a challenging dataset, named JD-landmark. Each image is manually annotated with 106-point landmarks. This dataset covers large variations on pose and expression, which brings a lot of difficulties to predict accurate landmarks. We hold a 106-point facial landmark localization competition 1 on this dataset in conjunction with IEEE International Conference on Multimedia and Expo (ICME) 2019. The purpose of this competition is to discover effective and robust facial landmark localization approaches.

The Visual Object Tracking challenge VOT2019 is the seventh annual tracker benchmarking activity organized by the VOT initiative. Results of 81 trackers are presented; many are state-of-the-art trackers published at major computer vision conferences or in journals in the recent years. The evaluation included the standard VOT and other popular methodologies for short-term tracking analysis as well as the standard VOT methodology for long-term tracking analysis. The VOT2019 challenge was composed of five challenges focusing on different tracking domains: (i) VOTST2019 challenge focused on short-term tracking in RGB, (ii) VOT-RT2019 challenge focused on "real-time" shortterm tracking in RGB, (iii) VOT-LT2019 focused on longterm tracking namely coping with target disappearance and reappearance. Two new challenges have been introduced: (iv) VOT-RGBT2019 challenge focused on short-term tracking in RGB and thermal imagery and (v) VOT-RGBD2019 challenge focused on long-term tracking in RGB and depth imagery. The VOT-ST2019, VOT-RT2019 and VOT-LT2019 datasets were refreshed while new datasets were introduced for VOT-RGBT2019 and VOT-RGBD2019. The VOT toolkit has been updated to support both standard shortterm, long-term tracking and tracking with multi-channel imagery. 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.

Model update lies at the heart of object tracking.Generally, model update is formulated as an online learning problem where a target model is learned over the online training dataset. Our key innovation is to \emph{learn the online learning algorithm itself using large number of offline videos}, i.e., \emph{learning to update}. The learned updater takes as input the online training dataset and outputs an updated target model. As a first attempt, we design the learned updater based on recurrent neural networks (RNNs) and demonstrate its application in a template-based tracker and a correlation filter-based tracker. Our learned updater consistently improves the base trackers and runs faster than realtime on GPU while requiring small memory footprint during testing. Experiments on standard benchmarks demonstrate that our learned updater outperforms commonly used update baselines including the efficient exponential moving average (EMA)-based update and the well-designed stochastic gradient descent (SGD)-based update. Equipped with our learned updater, the template-based tracker achieves state-of-the-art performance among realtime trackers on GPU.

The Fisher Vector (FV) is a very successful image representing method, which has achieved the state-of-the-art performance on scene classification. It concatenates the gradient of parameters in generative model as the image representation, which takes the advantage of generative and discriminative models. Using Gaussian mixture model (GMM) as the dictionary model, it can be regarded as an extension of the Bag-of-Words (BoW). But using unsupervised GMM to learn the dictionary makes a great loss for the information of image labels, which counts a lot for discrimination. To address the problem, we propose a novel strategy named Weakly-Supervised Gaussian Dictionary for Fisher Vector (WSGD-FV) to get the image representation. Specifically, we first use the weakly-supervised method to learn the Gaussian words, and then we combine these words to a Gaussian dictionary as the probability density function, so we can use this function to generate the FV. Our method is shown to get much better performance than the conventional FV for scene classification.