We introduce Lumina-Image 2.0, an advanced text-to-image generation framework that achieves significant progress compared to previous work, Lumina-Next. Lumina-Image 2.0 is built upon two key principles: (1) Unification - it adopts a unified architecture (Unified Next-DiT) that treats text and image tokens as a joint sequence, enabling natural cross-modal interactions and allowing seamless task expansion. Besides, since high-quality captioners can provide semantically well-aligned text-image training pairs, we introduce a unified captioning system, Unified Captioner (UniCap), specifically designed for T2I generation tasks. UniCap excels at generating comprehensive and accurate captions, accelerating convergence and enhancing prompt adherence. (2) Efficiency - to improve the efficiency of our proposed model, we develop multi-stage progressive training strategies and introduce inference acceleration techniques without compromising image quality. Extensive evaluations on academic benchmarks and public text-to-image arenas show that Lumina-Image 2.0 delivers strong performances even with only 2.6B parameters, highlighting its scalability and design efficiency. We have released our training details, code, and models at https://github.com/Alpha-VLLM/Lumina-Image-2.0.

Sora unveils the potential of scaling Diffusion Transformer (DiT) for generating photorealistic images and videos at arbitrary resolutions, aspect ratios, and durations, yet it still lacks sufficient implementation details. In this paper, we introduce the Lumina-T2X family -- a series of Flow-based Large Diffusion Transformers (Flag-DiT) equipped with zero-initialized attention, as a simple and scalable generative framework that can be adapted to various modalities, e.g., transforming noise into images, videos, multi-view 3D objects, or audio clips conditioned on text instructions. By tokenizing the latent spatial-temporal space and incorporating learnable placeholders such as |[nextline]| and |[nextframe]| tokens, Lumina-T2X seamlessly unifies the representations of different modalities across various spatial-temporal resolutions. Advanced techniques like RoPE, KQ-Norm, and flow matching enhance the stability, flexibility, and scalability of Flag-DiT, enabling models of Lumina-T2X to scale up to 7 billion parameters and extend the context window to 128K tokens. This is particularly beneficial for creating ultra-high-definition images with our Lumina-T2I model and long 720p videos with our Lumina-T2V model. Remarkably, Lumina-T2I, powered by a 5-billion-parameter Flag-DiT, requires only 35% of the training computational costs of a 600-million-parameter naive DiT (PixArt-alpha), indicating that increasing the number of parameters significantly accelerates convergence of generative models without compromising visual quality. Our further comprehensive analysis underscores Lumina-T2X's preliminary capability in resolution extrapolation, high-resolution editing, generating consistent 3D views, and synthesizing videos with seamless transitions. All code and checkpoints of Lumina-T2X are released at https://github.com/Alpha-VLLM/Lumina-T2X to further foster creativity, transparency, and diversity in the generative AI community.

Achieving generalization for deep learning models has usually suffered from the bottleneck of annotated sample scarcity. As a common way of tackling this issue, few-shot learning focuses on "episodes", i.e., sampled tasks that help the model acquire generalizable knowledge onto unseen categories - better the episodes, the higher a model's generalisability. Despite extensive research, the characteristics of episodes and their potential effects are relatively less explored. A recent paper discussed that different episodes exhibit different prediction difficulties, and coined a new metric "hardness" to quantify episodes, which however is too wide-range for an arbitrary dataset and thus remains impractical for realistic applications. In this paper therefore, we for the first time conduct an algebraic analysis of the critical factors influencing episode hardness supported by experimental demonstrations, that reveal episode hardness to largely depend on classes within an episode, and importantly propose an efficient pre-sampling hardness assessment technique named Inverse-Fisher Discriminant Ratio (IFDR). This enables sampling hard episodes at the class level via class-level (CL) sampling scheme that drastically decreases quantification cost. Delving deeper, we also develop a variant called class-pair-level (CPL) sampling, which further reduces the sampling cost while guaranteeing the sampled distribution. Finally, comprehensive experiments conducted on benchmark datasets verify the efficacy of our proposed method.

Compositional Zero-Shot Learning (CZSL) aims to recognize novel compositions using knowledge learned from seen attribute-object compositions in the training set. Previous works mainly project an image and its corresponding composition into a common embedding space to measure their compatibility score. However, both attributes and objects share the visual representations learned above, leading the model to exploit spurious correlations and bias towards seen compositions. Instead, we reconsider CZSL as an out-of-distribution generalization problem. If an object is treated as a domain, we can learn object-invariant features to recognize attributes attached to any object reliably, and vice versa. Specifically, we propose an invariant feature learning framework to align different domains at the representation and gradient levels to capture the intrinsic characteristics associated with the tasks. To further facilitate and encourage the disentanglement of attributes and objects, we propose an "encoding-reshuffling-decoding" process to help the model avoid spurious correlations by randomly regrouping the disentangled features into synthetic features. Ultimately, our method improves generalization by learning to disentangle features that represent two independent factors of attributes and objects. Experiments demonstrate that the proposed method achieves state-of-the-art or competitive performance in both closed-world and open-world scenarios.

Despite significant progress, the shortage of labeled data and expert knowledge remains a challenge for Fine-grained Visual Classification (FGVC). Some multi-source approaches that incorporate additional modalities, such as sound or bounding boxes, show promise for data enrichment but introduce added complexity to data collection. In this paper, we pose the question: can multi-source capabilities be achieved solely with existing images? The answer, confirmed by a pilot study, is affirmative. By analyzing the probability distribution of model output with different resolutions image, we find that complementary information beneficial to FGVC exists among images of different resolutions. Although the classification accuracy of low-resolution images is lower than high-resolution images, it can provide additional information for high-resolution input images. We designed a naive baseline that uses mixed training of multi-resolution images. Through the experimental results of the baseline, we find that i) not all low-resolution images are beneficial, and ii) adaptively selecting low-resolution images is what we need. Therefore, we proposed a meta-learning-based adaptive “resolution” pooling layer. Through the pooling operation, the features of low-resolution images are obtained from high-resolution images, and the most appropriate complementary features are selected for the features of high-resolution images through the gating mechanism, which enables the model to fully and autonomously exploit the complementary information. Experimental results on three FGVC datasets validate the effectiveness of our proposed method. Our code is available athttps://github.com/PRIS-CV/Adaptive-Multi-Resolution-Feature-Fusion.

Model customization mitigates the issues of inadequate performance, resource wastage, and privacy risks associated with using general-purpose models in specialized domains and well-defined tasks. However, achieving customization at a low annotation cost still poses a challenge. Existing domain adaptation research has addressed cases where all customized classes are present in the labeled database, yet scenarios involving customer-specific classes are still unresolved. Therefore, this paper proposes a novel Class-Customized Domain Adaptation (CCDA) method, addressing the latter scenario with just one additional annotation for each customer-specific class. CCDA adopts the classic adaptation training framework and comprises two innovative techniques. Firstly, to ensure the shared class knowledge from the database and the private class knowledge from additional annotations are transferred and propagated to the correct regions within the target domain, we design the partial-feature alignment strategy, based on the mechanical properties of feature alignment. Second, we propose soft-balanced sampling to tackle the long-tail distribution problem in labeled data, preventing the model from overfitting to the labeled samples of customer-specific classes. The effectiveness of CCDA has been validated across 48 tasks simulated on domain adaptation benchmarks and two real-world customization scenarios, consistently showing excellent performance. Additionally, extensive analytical experiments illustrate the contributions of two innovative techniques. The code is available at https://github.com/CHEN-kx/ClassCustomizedDA.

High-resolution image generation with Generative Artificial Intelligence (GenAl) has immense potential but, due to the enormous capital investment required for training, it is increasimgly centralised to a few large corporations, and hidden behind paywalls. This paper aims to democratise high-resolution GenAl by advancing the frontier of high-resolution generation while remaining accessible to a broad audience. We demonstrate that existing Latent Diffusion Models (LDMs) possess untapped potential for higher-resolution image generation. Our novel DemoFusion framework seamlessly extends open-source GenAl models, employing Progressive Upscaling, Skip Residual, and Di-lated Sampling mechanisms to achieve higher-resolution image generation. The progressive nature of DemoFusion requires more passes, but the intermediate results can serve as “previews”, facilitating rapid prompt iteration.

Lumina-T2X is a nascent family of Flow-based Large Diffusion Transformers (Flag-DiT) that establishes a unified framework for transforming noise into various modalities, such as images and videos, conditioned on text instructions. Despite its promising capabilities, Lumina-T2X still encounters challenges including training instability, slow inference, and extrapolation artifacts. In this paper, we present Lumina-Next, an improved version of Lumina-T2X, showcasing stronger generation performance with increased training and inference efficiency. We begin with a comprehensive analysis of the Flag-DiT architecture and identify several suboptimal components, which we address by introducing the Next-DiT architecture with 3D RoPE and sandwich normalizations. To enable better resolution extrapolation, we thoroughly compare different context extrapolation methods applied to text-to-image generation with 3D RoPE, and propose Frequency- and Time-Aware Scaled RoPE tailored for diffusion transformers. Additionally, we introduce a sigmoid time discretization schedule for diffusion sampling, which achieves high-quality generation in 5-10 steps combined with higher-order ODE solvers. Thanks to these improvements, Lumina-Next not only improves the basic text-to-image generation but also demonstrates superior resolution extrapolation capabilities as well as multilingual generation using decoder-based LLMs as the text encoder, all in a zero-shot manner. To further validate Lumina-Next as a versatile generative framework, we instantiate it on diverse tasks including visual recognition, multi-views, audio, music, and point cloud generation, showcasing strong performance across these domains. By releasing all codes and model weights at https://github.com/Alpha-VLLM/Lumina-T2X, we aim to advance the development of next-generation generative AI capable of universal modeling.

Despite great strides made on fine-grained visual classification (FGVC), current methods are still heavily reliant on fully-supervised paradigms where ample expert labels are called for. Semi-supervised learning (SSL) techniques, acquiring knowledge from unlabeled data, provide a considerable means forward and have shown great promise for coarse-grained problems. However, exiting SSL paradigms mostly assume in-category (i.e., category-aligned) unlabeled data, which hinders their effectiveness when re-proposed on FGVC. In this paper, we put forward a novel design specifically aimed at making out-of-category data work for semi-supervised FGVC. We work off an important assumption that all fine-grained categories naturally follow a hierarchical structure (e.g., the phylogenetic tree of "Aves" that covers all bird species). It follows that, instead of operating on individual samples, we can instead predict sample relations within this tree structure as the optimization goal of SSL. Beyond this, we further introduced two strategies uniquely brought by these tree structures to achieve inter-sample consistency regularization and reliable pseudo-relation. Our experimental results reveal that (i) the proposed method yields good robustness against out-of-category data, and (ii) it can be equipped with prior arts, boosting their performance thus yielding state-of-the-art results.

Generating both diverse and accurate descriptions is an essential goal in the audio captioning task. Traditional methods mainly focus on improving the accuracy of the generated captions but ignore their diversity. In contrast, recent methods have considered generating diverse captions for a given audio clip, but with the potential trade-off in caption accuracy. In this work, we propose a new diverse audio captioning method based on a variational autoencoder structure, dubbed AC-VAE, aiming to achieve a better trade-off between the diversity and accuracy of the generated captions. To improve diversity, AC-VAE learns the latent word distribution at each location based on contextual information. To uphold accuracy, AC-VAE incorporates an autoregressive prior module and a global constraint module, which enable precise modeling of word distribution and encourage semantic consistency of captions at the sentence level. We evaluate the proposed AC-VAE on the Clotho dataset. Experimental results show that AC-VAE achieves a better trade-off between diversity and accuracy compared to the state-of-the-art methods. The code is publicly available at https://github.com/XinMing0411/AC-VAE

Current fine-grained visual classification (FGVC) models are isolated. In practice, we first need to identify the coarse-grained label of an object, then select the corresponding FGVC model for recognition. This hinders the application of FGVC algorithms in real-life scenarios. In this paper, we propose an erudite FGVC model jointly trained by several different datasets11In this paper, different datasets mean different fine-grained visual classification datasets., which can efficiently and accurately predict an object's fine-grained label across the combined label space. We found through a pilot study that positive and negative transfers co-occur when different datasets are mixed for training, i.e., the knowledge from other datasets is not always useful. Therefore, we first propose a feature disentanglement module and a feature re-fusion module to reduce negative transfer and boost positive transfer between different datasets. In detail, we reduce negative transfer by decoupling the deep features through many dataset-specific feature extractors. Subsequently, these are channel-wise re-fused to facilitate positive transfer. Finally, we propose a meta-learning based dataset-agnostic spatial attention layer to take full advantage of the multi-dataset training data, given that localisation is dataset-agnostic between different datasets. Experimental results across 11 different mixed-datasets built on four different FGVC datasets demonstrate the effectiveness of the proposed method. Furthermore, the proposed method can be easily combined with existing FGVC methods to obtain state-of-the-art results. Our code is available at https://github.com/PRIS-CV/An-Erudite-FGVC-Model.

Although machines have surpassed humans on visual recognition problems, they are still limited to providing closed-set answers. Unlike machines, humans can cognize novel categories at the first observation. Novel category discovery (NCD) techniques, transferring knowledge from seen categories to distinguish unseen categories, aim to bridge the gap. However, current NCD methods assume a transductive learning and offline inference paradigm, which restricts them to a predefined query set and renders them unable to deliver instant feedback. In this paper, we study on-the-fly category discovery (OCD) aimed at making the model instantaneously aware of novel category samples (i.e., enabling inductive learning and streaming inference). We first design a hash coding-based expandable recognition model as a practical baseline. Afterwards, noticing the sensitivity of hash codes to intra-category variance, we further propose a novel Sign-Magnitude dIsentangLEment (SMILE) architecture to alleviate the disturbance it brings. Our experimental results demonstrate the superiority of SMILE against our baseline model and prior art. Our code is available at https://github.com/PRIS-CV/On-the-fly-Category-Discovery.

Despite the remarkable progress of Fine-grained visual classification (FGVC) with years of history, it is still limited to recognizing 2D images. Recognizing objects in the physical world (i.e., 3D environment) poses a unique challenge – discriminative information is not only present in visible local regions but also in other unseen views. Therefore, in addition to finding the distinguishable part from the current view, efficient and accurate recognition requires inferring the critical perspective with minimal glances. E.g., a person might recognize a "Ford sedan" with a glance at its side and then know that looking at the front can help tell which model it is. In this paper, towards FGVC in the real physical world, we put forward the problem of multi-view active fine-grained visual recognition (MAFR) and complete this study in three steps: (i) a multi-view, fine-grained vehicle dataset is collected as the testbed, (ii) a pilot experiment is designed to validate the need and research value of MAFR, (iii) a policy-gradient-based framework along with a dynamic exiting strategy is proposed to achieve efficient recognition with active view selection. Our comprehensive experiments demonstrate that the proposed method outperforms previous multi-view recognition works and can extend existing state-of-the-art FGVC methods and advanced neural networks to become "FGVC experts" in the 3D environment. Our code is available at https://github.com/PRIS-CV/MAFR.

Although existing few-shot learning works yield promising results for in-domain queries, they still suffer from weak cross-domain generalization. Limited support data requires effective knowledge transfer, but domain-shift makes this harder. Towards this emerging challenge, researchers improved adaptation by introducing task-specific parameters, which are directly optimized and estimated for each task. However, adding a fixed number of additional parameters fails to consider the diverse domain shifts between target tasks and the source domain, limiting efficacy. In this paper, we first observe the dependence of task-specific parameter configuration on the target task. Abundant task-specific parameters may over-fit, and insufficient task-specific parameters may result in under-adaptation – but the optimal task-specific configuration varies for different test tasks. Based on these findings, we propose the Task-aware Adaptive Network (TA2-Net), which is trained by reinforcement learning to adaptively estimate the optimal task-specific parameter configuration for each test task. It learns, for example, that tasks with significant domain-shift usually have a larger need for task-specific parameters for adaptation. We evaluate our model on Meta-dataset. Empirical results show that our model outperforms existing state-of-the-art methods. Our code is available at https://github.com/PRIS-CV/TA2-Net.

Weakly-supervised referring expression grounding (REG) aims at locating the image region described by a query sentence, where the mapping between the referential region and query is not available during the training stage. Noticing the significant gap between the fully- and weakly-supervised approaches, we develop a Self-Enhanced Training(SET) framework in this paper. Specifically, we first train the network under a weakly-supervised setting. Then, the model outputs are collected and filtered according to the confidence score and serve as pseudo-labels. Finally, with the help of these pseudo-labels, we tune the model under a fully-supervised setting. The SET framework provides a simple way of generating pseudo-labels that build a bridge between weak and full supervision. Experimental results demonstrate that model trained through our SET framework outperforms existing traditional methods on RefCOCO, RefCOCO+, and RefCOCOg datasets. The code is available at https://github.com/HTDL98/SET-framework.

As powerful as fine-grained visual classification (FGVC) is, responding your query with a bird name of “Whip-poor-will” or “Mallard” probably does not make much sense. This however commonly accepted in the literature, underlines a fundamental question interfacing AI and human – what constitutes transferable knowledge for human to learn from AI? This paper sets out to answer this very question using FGVC as a test bed. Specifically, we envisage a scenario where a trained FGVC model (the AI expert) functions as a knowledge provider in enabling average people (you and me) to become better domain experts ourselves,i.e.,those capable in distinguishing between “Whip-poor-will” and “Mallard”. Fig. 1 lays out our approach in answering this question. Assuming an AI expert trained using expert human labels, we ask (i) what is the best transferable knowledge we can extract from AI, and (ii) what is the most practical means to measure the gains in expertise given that knowledge? On the former, we propose to represent knowledge as highly discriminative visual regions that are expert-exclusive. For that, we devise a multi-stage learning framework, which starts with modelling visual attention of domain experts and novices separately, before discriminatively distilling their differences to acquire those exclusive to experts. For the latter, we simulate the evaluation process as a book guide to best accommodate the learning practice of that is accustomed to humans. A comprehensive human study of 15,000 trials shows our method is able to consistently improve people of divergent bird expertise to recognise once unrecognisable birds. To counter the lack of reproducibility of perceptual studies, and in turn to make a sustainable direction out of our “AI for Human” effort, we further propose a quantitative metric, namely Transferable Effective Model Attention (TEMI). TEMI acts as a crude but benchmarkable metric to replace large-scale human studies, and therefore allows future efforts in this direction to be comparable to ours. We attest to the integrity of TEMI by (i) empirically showing a strong correlation between TEMI scores and raw human study data, and (ii) its expected behaviour holds for a large body of attention models. Last but not least, our approach also leads to improved FGVC performance in the conventional benchmarking sense, when the extracted knowledge defined is utilised as means to achieve discriminative localisation. Codes and all details on the human study are available at:https://github.com/PRIS-CV/Making-a-Bird-AI-Expert-Work-for-You-and-Me.

Audio captioning aims at describing the content of audio clips with human language. Due to the ambiguity of audio content, different people may perceive the same audio clip differently, resulting in caption disparities (i.e., the same audio clip may be described by several captions with diverse semantics). In the literature, the one-to-many strategy is often employed to train the audio captioning models, where a related caption is randomly selected as the optimization target for each audio clip at each training iteration. However, we observe that this can lead to significant variations during the optimization process and adversely affect the performance of the model. In this paper, we address this issue by proposing an audio captioning method, named ACTUAL (Audio Captioning with capTion featUre spAce reguLarization). ACTUAL involves a two-stage training process: (i) in the first stage, we use contrastive learning to construct a proxy feature space where the similarities between captions at the audio level are explored, and (ii) in the second stage, the proxy feature space is utilized as additional supervision to improve the optimization of the model in a more stable direction. We conduct extensive experiments to demonstrate the effectiveness of the proposed ACTUAL method. The results show that proxy caption embedding can significantly improve the performance of the baseline model and the proposed ACTUAL method offers competitive performance on two datasets compared to state-of-the-art methods. The code is publicly available athttps://github.com/PRIS-CV/Caption-Feature-Space-Regularization.

Adaptation to out-of-distribution data is a meta-challenge for all statistical learning algorithms that strongly rely on the i.i.d. assumption. It leads to unavoidable labor costs and confidence crises in realistic applications. For that, domain generalization aims at mining domain-irrelevant knowledge from multiple source domains that can generalize to unseen target domains. In this paper, by leveraging the frequency domain of an image, we uniquely work with two key observations: (i) the high-frequency information of an image depicts object edge structure, which preserves high-level semantic information of the object is naturally consistent across different domains, and (ii) the low-frequency component retains object smooth structure, while this information is susceptible to domain shifts. Motivated by the above observations, we introduce (i) an encoder-decoder structure to disentangle high- and low-frequency features of an image, (ii) an information interaction mechanism to ensure the helpful knowledge from both two parts can cooperate effectively, and (iii) a novel data augmentation technique that works on the frequency domain to encourage the robustness of frequency-wise feature disentangling. The proposed method obtains state-of-the-art performance on three widely used domain generalization benchmarks (Digit-DG, Office-Home, and PACS).

In recent years, fine-grained visual classification (FGVC) algorithms have achieved excellent performance across a variety of datasets. However, it is still rare to see these algorithms applied in daily life. The main reasons for this are i) the algorithms are developed based on different design guidelines and cannot be deployed in the same environment; ii) there is not a simple and efficient platform to present the algorithm's results to the user - the accuracy is meaningless to the users. To address the above problem, we built a complex scenario-oriented fine-grained visual classification platform. The platform consists of a PyTorch-based fine-grained visual recognition algorithm library (FGL) and a WeChat applet-based user interaction module (WEM). We can quickly develop new algorithms or readily apply existing algorithms in the same environment through FGL. Driven by FGL, the WEM enables users to achieve fine-grained recognition of complex scenes interactively. In addition to showing the user the fine-grained labels of objects, we will also show how the model makes decisions to help the user master the ability to recognise the fine-grained object so that everyone can become a domain expert. A video demo shows an example of the proposed platform in a real-world scenario: https://reurl.cc/rRZE7O.

Enabling machines to process multi-modal information and understand the real physical world is an important step to achieving free human-machine conversation. However, previous human-machine conversation systems are mostly limited to single-modal (e.g., chat robot), single-round (e.g., visual Q & A), and static visual information (e.g., visual dialogue). To address the above problem, we develop a multi-modal human-machine conversation system for specific application scenarios. The system includes two modules: (i) an interactive visual grounding module that can actively disambiguate user's queries, and (ii) an interactive fine-grained recognition module that can model objects in the 3D environment and actively ask for missing visual information. A video demo of our system under the automobile sales scenario can be found here11https://drive.google.com/file/d/1IfBsMKq55ryLOZchIT6G-R5CyWnC3Skiew?usp=sharing.

In contrast to traditional fine-grained visual clas-sification, multi-granularity visual classification is no longer limited to identifying the different sub-classes belonging to the same super-class (e.g., bird species, cars, and aircraft models). Instead, it gives a sequence of labels from coarse to fine (e.g., Passeriformes → Corvidae → Fish Crow), which is more convenient in practice. The key to solving this task is how to use the relationships between the different levels of labels to learn feature representations that contain different levels of granularity. Interestingly, the feature pyramid structure naturally implies different granularity of feature representation, with the shallow layers representing coarse-grained features and the deep layers representing fine-grained features. Therefore, in this paper, we exploit this property of the feature pyramid structure to decouple features and obtain feature representations corre-sponding to different granularities. Specifically, we use shallow features for coarse-grained classification and deep features for fine-grained classification. In addition, to enable fine-grained features to enhance the coarse-grained classification, we propose a feature reinforcement module based on the feature pyramid structure, where deep features are first upsampled and then combined with shallow features to make decisions. Experimental results on three widely used fine-grained image classification datasets such as CUB-200-2011, Stanford Cars, and FGVC-Aircraft validate the method's effectiveness. Code available at https://github.com/PRIS-CV/CGVC.

Fine-grained visual classification (FGVC) is much more challenging than traditional classification tasks due to the inherently subtle intra-class object variations. Recent works are mainly part-driven (either explicitly or implicitly), with the assumption that fine-grained information naturally rests within the parts. In this paper, we take a different stance, and show that part operations are not strictly necessary - the key lies with encouraging the network to learn at different granularities and progressively fusing multi-granularity features together. In particular, we propose: (i) a progressive training strategy that effectively fuses features from different granularities, and (ii) a consistent block convolution that encourages the network to learn the category-consistent features at specific granularities. We evaluate on several standard FGVC benchmark datasets, and demonstrate the proposed method consistently outperforms existing alternatives or delivers competitive results. Codes are available at https://github.com/PRIS-CV/PMG-V2.

Metric-based methods achieve promising performance on few-shot classification by learning clusters on support samples and generating shared decision boundaries for query samples. However, existing methods ignore the inaccurate class center approximation introduced by the limited number of support samples, which consequently leads to biased inference. Therefore, in this paper, we propose to reduce the approximation error by class center calibration. Specifically, we introduce the so-called Pair-wise Similarity Module (PSM) to generate calibrated class centers adapted to the query sample by capturing the semantic correlations between the support and the query samples, as well as enhancing the discriminative regions on support representation. It is worth noting that the proposed PSM is a simple plug-and-play module and can be inserted into most metric-based few-shot learning models. Through extensive experiments in metric-based models, we demonstrate that the module significantly improves the performance of conventional few-shot classification methods on four few-shot image classification benchmark datasets. Codes are available at: https://github.com/PRIS-CV/Pair-wise-Similarity-module.

Fine-grained visual classification (FGVC) aims to distinguish the sub-classes of the same category and its essential solution is to mine the subtle and discriminative regions. Convolution neural networks (CNNs), which employ the cross entropy loss (CE-loss) as the loss function, show poor performance since the model can only learn the most discriminative part and ignore other meaningful regions. Some existing works try to solve this problem by mining more discriminative regions by some detection techniques or attention mechanisms. However, most of them will meet the background noise problem when trying to find more discriminative regions. In this paper, we address it in a knowledge transfer learning manner. Multiple models are trained one by one, and all previously trained models are regarded as teacher models to supervise the training of the current one. Specifically, a orthogonal loss (OR-loss) is proposed to encourage the network to find diverse and meaningful regions. In addition, the first model is trained with only CE-Loss. Finally, all models’ outputs with complementary knowledge are combined together for the final prediction result. We demonstrate the superiority of the proposed method and obtain state-of-the-art (SOTA) performances on three popular FGVC datasets.