This paper introduces a robot active task cognition framework for Situation-Aware Task Planning (SATP), leveraging visual scene understanding to generate action sequences. By integrating object knowledge, user preferences, and Large Language Models (LLMs), SATP interprets the robot’s current visual perception, and creates procedural actions that align with what the robot “sees”. Diverging from conventional methods requiring explicit verbal commands, our SATP framework autonomously performs task cognition, actively formulating robot-executable action sequences directly from visual input. Initially, a novel approach for describing the visual scene is presented, enabling the robot to grasp detailed object-level properties and inter-object relationships based on its observations. Building on this, a knowledge base for active task cognition is constructed using ontology technology. Furthermore, we develop a two-stage dual-feedback task planner, ReProg+, powered by LLMs, specifically designed for situation-aware task planning grounded in visual data. The efficacy, reliability, and advantages of our solution are thoroughly validated in real-world visual scenarios. Additionally, SATP has been tested with a real robot, with results confirming the feasibility and effectiveness of our approach.

Online Knowledge Distillation (OKD) methods represent a streamlined, one-stage distillation training process that obviates the necessity of transferring knowledge from a pretrained teacher network to a more compact student network. In contrast to existing logits-based OKD methods, this paper presents an innovative approach to leverage intermediate spatial representations. Our analysis of the intermediate features from both teacher and student models reveals two pivotal insights: (1) the similar features between students and teachers are predominantly focused on the foreground objects. (2) teacher models emphasize foreground objects more than students. Building on these findings, we propose Asymmetric Decision-Making (ADM) to enhance feature consensus learning for student models while continuously promoting feature diversity in teacher models. Specifically, Consensus Learning for student models prioritizes spatial features with high consensus relative to teacher models. Conversely, Divergence Learning for teacher models highlights spatial features with lower similarity compared to student models, indicating superior performance by teacher models in these regions. Consequently, ADM facilitates the student models to catch up with the feature learning process of the teacher models. Extensive experiments demonstrate that ADM consistently surpasses existing OKD methods across various online knowledge distillation settings and also achieves superior results when transferred to offline knowledge distillation, semantic segmentation and diffusion distillation tasks.

Most existing distillation methods ignore the flexible role of the temperature in the loss function and fix it as a hyper-parameter that can be decided by an inefficient grid search. In general, the temperature controls the discrepancy between two distributions and can faithfully determine the difficulty level of the distillation task. Keeping a constant temperature, i.e., a fixed level of task difficulty, is usually sub-optimal for a growing student during its progressive learning stages. In this paper, we propose a simple curriculum-based technique, termed Curriculum Temperature for Knowledge Distillation (CTKD), which controls the task difficulty level during the student's learning career through a dynamic and learnable temperature. Specifically, following an easy-to-hard curriculum, we gradually increase the distillation loss w.r.t. the temperature, leading to increased distillation difficulty in an adversarial manner. As an easy-to-use plug-in technique, CTKD can be seamlessly integrated into existing knowledge distillation frameworks and brings general improvements at a negligible additional computation cost. Extensive experiments on CIFAR-100, ImageNet-2012, and MS-COCO demonstrate the effectiveness of our method.

Semi-supervised learning (SSL) has attracted enormous attention due to its vast potential of mitigating the dependence on large labeled datasets. The latest methods (e.g., FixMatch) use a combination of consistency regularization and pseudo-labeling to achieve remarkable successes. However, these methods all suffer from the waste of complicated examples since all pseudo-labels have to be selected by a high threshold to filter out noisy ones. Hence, the examples with ambiguous predictions will not contribute to the training phase. For better leveraging all unlabeled examples, we propose two novel techniques: Entropy Meaning Loss (EML) and Adaptive Negative Learning (ANL). EML incorporates the prediction distribution of non-target classes into the optimization objective to avoid competition with target class, and thus generating more high-confidence predictions for selecting pseudo-label. ANL introduces the additional negative pseudo-label for all unlabeled data to leverage low-confidence examples. It adaptively allocates this label by dynamically evaluating the top-k performance of the model. EML and ANL do not introduce any additional parameter and hyperparameter. We integrate these techniques with FixMatch, and develop a simple yet powerful framework called FullMatch. Extensive experiments on several common SSL benchmarks (CIFAR-10/100, SVHN, STL-10 and ImageNet) demonstrate that FullMatch exceeds FixMatch by a large margin. Integrated with FlexMatch (an advanced FixMatch-based framework), we achieve state-of-the-art performance. Source code is available at https://github.com/megvii-research/FullMatch.

Knowledge distillation transfers knowledge from a large model to a small one via task and distillation losses. In this paper, we observe a trade-off between task and distillation losses, i.e., introducing distillation loss limits the convergence of task loss. We believe that the trade-off results from the insufficient optimization of distillation loss. The reason is: The teacher has a lower task loss than the student, and a lower distillation loss drives the student more similar to the teacher, then a better-converged task loss could be obtained. To break the trade-off, we propose the Distillation-Oriented Trainer (DOT). DOT separately considers gradients of task and distillation losses, then applies a larger momentum to distillation loss to accelerate its optimization. We empirically prove that DOT breaks the trade-off, i.e., both losses are sufficiently optimized. Extensive experiments validate the superiority of DOT. Notably, DOT achieves a +2.59% accuracy improvement on ImageNet-1k for the ResNet50-MobileNetV1 pair. Conclusively, DOT greatly benefits the student’s optimization properties in terms of loss convergence and model generalization. https://github.com/megvii-research/mdistiller.

Distilling knowledge from convolutional neural networks (CNNs) is a double-edged sword for vision transformers (ViTs). It boosts the performance since the image-friendly local-inductive bias of CNN helps ViT learn faster and better, but leading to two problems: (1) Network designs of CNN and ViT are completely different, which leads to different semantic levels of intermediate features, making spatial-wise knowledge transfer methods (e.g., feature mimicking) inefficient. (2) Distilling knowledge from CNN limits the network convergence in the later training period since ViT’s capability of integrating global information is suppressed by CNN’s local-inductive-bias supervision.To this end, we present Cumulative Spatial Knowledge Distillation (CSKD). CSKD distills spatial-wise knowledge to all patch tokens of ViT from the corresponding spatial responses of CNN, without introducing intermediate features. Furthermore, CSKD exploits a Cumulative Knowledge Fusion (CKF) module, which introduces the global response of CNN and increasingly emphasizes its importance during the training. Applying CKF leverages CNN’s local inductive bias in the early training period and gives full play to ViT’s global capability in the later one. Extensive experiments and analysis on ImageNet-1k and downstream datasets demonstrate the superiority of our CSKD. Code: https://github.com/Zzzzz1/CSKD

Fine-grained image classification is a challenging computer vision task where various species share similar visual appearances, resulting in misclassification if merely based on visual clues. Therefore, it is helpful to leverage additional information, e.g., the locations and dates for data shooting, which can be easily accessible but rarely exploited. In this paper, we first demonstrate that existing multimodal methods fuse multiple features only on a single dimension, which essentially has insufficient help in feature discrimination. To fully explore the potential of multimodal information, we propose a dynamic MLP on top of the image representation, which interacts with multimodal features at a higher and broader dimension. The dynamic MLP is an efficient structure parameterized by the learned embeddings of variable locations and dates. It can be regarded as an adaptive nonlinear projection for generating more discriminative image representations in visual tasks. To our best knowledge, it is the first attempt to explore the idea of dynamic networks to exploit multimodal information in fine-grained image classification tasks. Extensive experiments demonstrate the effectiveness of our method. The t-SNE algorithm visually indicates that our technique improves the recognizability of image representations that are visually similar but with different categories. Furthermore, among published works across multiple fine-grained datasets, dynamic MLP consistently achieves SOTA results11https://paperswithcode.com/dataset/inaturalist and takes third place in the iNaturalist challenge at FGVC822https://www.kaggle.com/c/inaturalist-2021/leaderboard. Code is available at httpsr//glthub.com/megvii-research/DynamicMLPForFinegrained.

State-of-the-art distillation methods are mainly based on distilling deep features from intermediate layers, while the significance of logit distillation is greatly overlooked. To provide a novel viewpoint to study logit distillation, we re-formulate the classical KD loss into two parts, i.e., target class knowledge distillation (TCKD) and non-target class knowledge distillation (NCKD). We empirically investigate and prove the effects of the two parts: TCKD transfers knowledge concerning the “difficulty” of training samples, while NCKD is the prominent reason why logit distillation works. More importantly, we reveal that the classical KD loss is a coupled formulation, which (1) suppresses the effectiveness of NCKD and (2) limits the flexibility to balance these two parts. To address these issues, we present Decoupled Knowledge Distillation (DKD), enabling TCKD and NCKD to play their roles more efficiently and flexibly. Compared with complex feature-based methods, our DKD achieves comparable or even better results and has better training efficiency on CIFAR-100, ImageNet, and MS-COCO datasets for image classification and object detection tasks. This paper proves the great potential of logit distillation, and we hope it will be helpful for future research. The code is available at https://github.com/megviiresearch/mdistiller.

Mix-based augmentation has been proven fundamental to the generalization of deep vision models. However, current augmentations only mix samples from the current data batch during training, which ignores the possible knowledge accumulated in the learning history. In this paper, we propose a recursive mixed-sample learning paradigm, termed ``RecursiveMix'' (RM), by exploring a novel training strategy that leverages the historical input-prediction-label triplets. More specifically, we iteratively resize the input image batch from the previous iteration and paste it into the current batch while their labels are fused proportionally to the area of the operated patches. Furthermore, a consistency loss is introduced to align the identical image semantics across the iterations, which helps the learning of scale-invariant feature representations. Based on ResNet-50, RM largely improves classification accuracy by $\sim$3.2% on CIFAR-100 and $\sim$2.8% on ImageNet with negligible extra computation/storage costs. In the downstream object detection task, the RM-pretrained model outperforms the baseline by 2.1 AP points and surpasses CutMix by 1.4 AP points under the ATSS detector on COCO. In semantic segmentation, RM also surpasses the baseline and CutMix by 1.9 and 1.1 mIoU points under UperNet on ADE20K, respectively. Codes and pretrained models are available at https://github.com/implus/RecursiveMix.

Multi-label image recognition has attracted considerable research attention and achieved great success in recent years. Capturing label correlations is an effective manner to advance the performance of multi-label image recognition. Two types of label correlations were principally studied, i.e., the spatial and semantic correlations. However, in the literature, previous methods considered only either of them. In this work, inspired by the great success of Transformer, we propose a plug-and-play module, named the Spatial and Semantic Transformers (SST), to simultaneously capture spatial and semantic correlations in multi-label images. Our proposal is mainly comprised of two independent transformers, aiming to capture the spatial and semantic correlations respectively. Specifically, our Spatial Transformer is designed to model the correlations between features from different spatial positions, while the Semantic Transformer is leveraged to capture the co-existence of labels without manually defined rules. Other than methodological contributions, we also prove that spatial and semantic correlations complement each other and deserve to be leveraged simultaneously in multi-label image recognition. Benefitting from the Transformer's ability to capture long-range correlations, our method remarkably outperforms state-of-the-art methods on four popular multi-label benchmark datasets. In addition, extensive ablation studies and visualizations are provided to validate the essential components of our method.

Despite significant progress of applying deep learning methods to the field of content-based image retrieval, there has not been a software library that covers these methods in a unified manner. In order to fill this gap, we introduce PyRetri, an open source library for deep learning based unsupervised image retrieval. The library encapsulates the retrieval process in several stages and provides functionality that covers various prominent methods for each stage. The idea underlying its design is to provide a unified platform for deep learning based image retrieval research, with high usability and extensibility. The project source code, with usage examples, sample data and pre-trained models are available at https://github.com/PyRetri/.

In this paper, we tackle a valuable yet very challenging visual recognition task, where the instances are within a subordinate category, and the target domain undergoes a shift with the source domain. This task, termed as cross-domain fine-grained recognition, relates closely to many real-life scenarios, e.g., recognizing retail products in storage racks by models trained with images collected in controlled environments. To deal with this problem, we design a new algorithm and propose a corresponding fine-grained domain adaptation dataset. Firstly, we propose a novel end-to-end CNN architecture that integrates two specialized modules: an adversarial module for domain alignment and a self-attention module for fine-grained recognition. The adversarial module is used to handle domain shift by gradually aligning the different domains with domain-level and class-level alignments, and strive to help the classifier learn with domain-invariant features generated by nets. The self-attention module is designed to capture discriminative image regions which are crucial for fine-grained visual recognition. Secondly, we collect a large-scale fine-grained domain adaptation dataset of retail products, which contains 52,011 images of 263 classes from 3 domains. Thirdly, we validate the effectiveness of our method on three datasets, showing that the proposed method can yield significant improvements over baseline methods on fine-grained datasets. Besides, we also evaluate the effectiveness of the self-attention module by performing visualization, which can capture the discriminative image regions in both source and target domains.

Traditional fine-grained image recognition is required to distinguish different subordinate categories (e.g., birds species) based on the visual cues beneath raw images. Due to both small inter-class variations and large intra-class variations, it is desirable to capture the subtle differences between these sub-categories, which is crucial but challenging for fine-grained recognition. Recently, language modality aggregation has been proved as a successful technique to improve visual recognition in the experience. In this paper, we introduce an end-to-end trainable Progressive Mask Attention (PMA) model for fine-grained recognition by leveraging both visual and language modalities. Our Bi-Modal PMA model can not only stage-by-stage capture the most discriminative part in the visual modality by our mask-based fashion, but also explore the out-of-visual-domain knowledge from the language modality in an interactional alignment paradigm. Specifically, at each stage, a self-attention module is proposed to attend to the key patch from images or text descriptions. Besides, a query-relational module is designed to seize the key words/phrases of texts and further bridge the connection between two modalities. Later, the learned representations of bi-modality from multiple stages are aggregated as the final features for recognition. Our Bi-Modal PMA model only needs raw images and raw text descriptions, without requiring bounding boxes/part annotations in images or key word annotations in texts. By conducting comprehensive experiments on fine-grained benchmark datasets, we demonstrate that the proposed method achieves superior performance over the competing baselines, on either vision and language bi-modality or single visual modality.

We study the problem of approximately counting matchings in hypergraphs of bounded maximum degree and maximum size of hyperedges. With an activity parameter λ , each matching M is assigned a weight λ | M | . The counting problem is formulated as computing a partition function that gives the sum of the weights of all matchings in a hypergraph. This problem unifies two extensively studied statistical physics models in approximate counting: the hardcore model (graph independent sets) and the monomer–dimer model (graph matchings). For this problem, the critical activity λ c = d d k ( d − 1 ) d + 1 is the threshold for the uniqueness of Gibbs measures on the infinite ( d + 1 ) -uniform ( k + 1 ) -regular hypertree. Consider hypergraphs of maximum degree at most k + 1 and maximum size of hyperedges at most d + 1 . We show that when λ < λ c , there is an FPTAS for computing the partition function; and when λ = λ c , there is a PTAS for computing the log-partition function. These algorithms are based on the decay of correlation (strong spatial mixing) property of Gibbs distributions. When λ > 2 λ c , there is no PRAS for the partition function or the log-partition function unless NP = RP. Towards obtaining a sharp transition of computational complexity of approximate counting, we study the local convergence from a sequence of finite hypergraphs to the infinite lattice with specified symmetry. We show a surprising connection between the local convergence and the reversibility of a natural random walk. This leads us to a barrier for the hardness result: The non-uniqueness of infinite Gibbs measure is not realizable by any finite gadgets.

We study the problem of approximately counting matchings in hypergraphs of bounded maximum degree and maximum size of hyperedges. With an activity parameter lambda, each matching M is assigned a weight lambda^{|M|}. The counting problem is formulated as computing a partition function that gives the sum of the weights of all matchings in a hypergraph. This problem unifies two extensively studied statistical physics models in approximate counting: the hardcore model (graph independent sets) and the monomer-dimer model (graph matchings). For this model, the critical activity lambda_c= (d^d)/(k (d-1)^{d+1}) is the threshold for the uniqueness of Gibbs measures on the infinite (d+1)-uniform (k+1)-regular hypertree. Consider hypergraphs of maximum degree at most k+1 and maximum size of hyperedges at most d+1. We show that when lambda < lambda_c, there is an FPTAS for computing the partition function; and when lambda = lambda_c, there is a PTAS for computing the log-partition function. These algorithms are based on the decay of correlation (strong spatial mixing) property of Gibbs distributions. When lambda > 2lambda_c, there is no PRAS for the partition function or the log-partition function unless NP=RP. Towards obtaining a sharp transition of computational complexity of approximate counting, we study the local convergence from a sequence of finite hypergraphs to the infinite lattice with specified symmetry. We show a surprising connection between the local convergence and the reversibility of a natural random walk. This leads us to a barrier for the hardness result: The non-uniqueness of infinite Gibbs measure is not realizable by any finite gadgets.