Few-shot fine-grained image classification (FS-FGIC) presents a significant challenge, requiring models to distinguish visually similar subclasses with limited labeled examples. Existing methods have critical limitations: metric-based methods lose spatial information and misalign local features, while reconstruction-based methods fail to utilize hierarchical feature information and lack mechanisms to focus on discriminative regions. We propose the Hierarchical Mask-enhanced Dual Reconstruction Network (HMDRN), which integrates dual-layer feature reconstruction with mask-enhanced feature processing to improve fine-grained classification. HMDRN incorporates a dual-layer feature reconstruction and fusion module that leverages complementary visual information from different network hierarchies. Through learnable fusion weights, the model balances high-level semantic representations from the last layer with mid-level structural details from the penultimate layer. Additionally, we design a spatial binary mask-enhanced transformer self-reconstruction module that processes query features through adaptive thresholding while maintaining complete support features, enhancing focus on discriminative regions while filtering background noise. Extensive experiments on three challenging fine-grained datasets demonstrate that HMDRN consistently outperforms state-of-the-art methods across Conv-4 and ResNet-12 backbone architectures. Comprehensive ablation studies validate the effectiveness of each proposed component, revealing that dual-layer reconstruction enhances inter-class discrimination while mask-enhanced transformation reduces intra-class variations. Visualization results provide evidence of HMDRN's superior feature reconstruction capabilities.
Given the severe challenges confronting the global growth security of economic crops, precise identification and prevention of plant diseases has emerged as a critical issue in artificial intelligence-enabled agricultural technology. To address the technical challenges in plant disease recognition, including small-sample learning, leaf occlusion, illumination variations, and high inter-class similarity, this study innovatively proposes a Dynamic Dual-Stream Fusion Network (DS_FusionNet). The network integrates a dual-backbone architecture, deformable dynamic fusion modules, and bidirectional knowledge distillation strategy, significantly enhancing recognition accuracy. Experimental results demonstrate that DS_FusionNet achieves classification accuracies exceeding 90% using only 10% of the PlantDisease and CIFAR-10 datasets, while maintaining 85% accuracy on the complex PlantWild dataset, exhibiting exceptional generalization capabilities. This research not only provides novel technical insights for fine-grained image classification but also establishes a robust foundation for precise identification and management of agricultural diseases.
Reinforcement Fine-Tuning (RFT) in Large Reasoning Models like OpenAI o1 learns from feedback on its answers, which is especially useful in applications when fine-tuning data is scarce. Recent open-source work like DeepSeek-R1 demonstrates that reinforcement learning with verifiable reward is one key direction in reproducing o1. While the R1-style model has demonstrated success in language models, its application in multi-modal domains remains under-explored. This work introduces Visual Reinforcement Fine-Tuning (Visual-RFT), which further extends the application areas of RFT on visual tasks. Specifically, Visual-RFT first uses Large Vision-Language Models (LVLMs) to generate multiple responses containing reasoning tokens and final answers for each input, and then uses our proposed visual perception verifiable reward functions to update the model via the policy optimization algorithm such as Group Relative Policy Optimization (GRPO). We design different verifiable reward functions for different perception tasks, such as the Intersection over Union (IoU) reward for object detection. Experimental results on fine-grained image classification, few-shot object detection, reasoning grounding, as well as open-vocabulary object detection benchmarks show the competitive performance and advanced generalization ability of Visual-RFT compared with Supervised Fine-tuning (SFT). For example, Visual-RFT improves accuracy by $24.3\%$ over the baseline in one-shot fine-grained image classification with around 100 samples. In few-shot object detection, Visual-RFT also exceeds the baseline by $21.9$ on COCO's two-shot setting and $15.4$ on LVIS. Our Visual-RFT represents a paradigm shift in fine-tuning LVLMs, offering a data-efficient, reward-driven approach that enhances reasoning and adaptability for domain-specific tasks.
Low-resolution fine-grained image classification has recently made significant progress, largely thanks to the super-resolution techniques and knowledge distillation methods. However, these approaches lead to an exponential increase in the number of parameters and computational complexity of models. In order to solve this problem, in this letter, we propose a Vision Mamba Distillation (ViMD) approach to enhance the effectiveness and efficiency of low-resolution fine-grained image classification. Concretely, a lightweight super-resolution vision Mamba classification network (SRVM-Net) is proposed to improve its capability for extracting visual features by redesigning the classification sub-network with Mamba modeling. Moreover, we design a novel multi-level Mamba knowledge distillation loss boosting the performance, which can transfer prior knowledge obtained from a High-resolution Vision Mamba classification Network (HRVM-Net) as a teacher into the proposed SRVM-Net as a student. Extensive experiments on seven public fine-grained classification datasets related to benchmarks confirm our ViMD achieves a new state-of-the-art performance. While having higher accuracy, ViMD outperforms similar methods with fewer parameters and FLOPs, which is more suitable for embedded device applications. Code is available at https://github.com/boa2004plaust/ViMD.
Large-scale, volunteer-collected datasets of community-identified natural world imagery like iNaturalist have enabled marked performance gains for fine-grained visual classification of species using machine learning methods. However, such data -- sometimes referred to as citizen science data -- are opportunistic and lack a structured sampling strategy. This volunteer-collected biodiversity data contains geographic, temporal, taxonomic, observers, and sociopolitical biases that can have significant effects on biodiversity model performance, but whose impacts are unclear for fine-grained species recognition performance. Here we introduce Diversity Shift (DivShift), a framework for quantifying the effects of domain-specific distribution shifts on machine learning model performance. To diagnose the performance effects of biases specific to volunteer-collected biodiversity data, we also introduce DivShift - North American West Coast (DivShift-NAWC), a curated dataset of almost 7.5 million iNaturalist images across the western coast of North America partitioned across five types of expert-verified bias. We compare species recognition performance across these bias partitions using a diverse variety of species- and ecosystem-focused accuracy metrics. We observe that these biases confound model performance less than expected from the underlying label distribution shift, and that more data leads to better model performance but the magnitude of these improvements are bias-specific. These findings imply that while the structure within natural world images provides generalization improvements for biodiversity monitoring tasks, the biases present in volunteer-collected biodiversity data can also affect model performance; thus these models should be used with caution in downstream biodiversity monitoring tasks.
Although it is traditionally believed that lossy image compression, such as JPEG compression, has a negative impact on the performance of deep neural networks (DNNs), it is shown by recent works that well-crafted JPEG compression can actually improve the performance of deep learning (DL). Inspired by this, we propose JPEG-DL, a novel DL framework that prepends any underlying DNN architecture with a trainable JPEG compression layer. To make the quantization operation in JPEG compression trainable, a new differentiable soft quantizer is employed at the JPEG layer, and then the quantization operation and underlying DNN are jointly trained. Extensive experiments show that in comparison with the standard DL, JPEG-DL delivers significant accuracy improvements across various datasets and model architectures while enhancing robustness against adversarial attacks. Particularly, on some fine-grained image classification datasets, JPEG-DL can increase prediction accuracy by as much as 20.9%. Our code is available on https://github.com/JpegInspiredDl/JPEG-Inspired-DL.git.
Accurately predicting the geographic ranges of species is crucial for assisting conservation efforts. Traditionally, range maps were manually created by experts. However, species distribution models (SDMs) and, more recently, deep learning-based variants offer a potential automated alternative. Deep learning-based SDMs generate a continuous probability representing the predicted presence of a species at a given location, which must be binarized by setting per-species thresholds to obtain binary range maps. However, selecting appropriate per-species thresholds to binarize these predictions is non-trivial as different species can require distinct thresholds. In this work, we evaluate different approaches for automatically identifying the best thresholds for binarizing range maps using presence-only data. This includes approaches that require the generation of additional pseudo-absence data, along with ones that only require presence data. We also propose an extension of an existing presence-only technique that is more robust to outliers. We perform a detailed evaluation of different thresholding techniques on the tasks of binary range estimation and large-scale fine-grained visual classification, and we demonstrate improved performance over existing pseudo-absence free approaches using our method.
Text recognition in the wild is an important technique for digital maps and urban scene understanding, in which the natural resembling properties between glyphs is one of the major reasons that lead to wrong recognition results. To address this challenge, we introduce two extremely fine-grained visual recognition benchmark datasets that contain very challenging resembling glyphs (characters/letters) in the wild to be distinguished. Moreover, we propose a simple yet effective two-stage contrastive learning approach to the extremely fine-grained recognition task of resembling glyphs discrimination. In the first stage, we utilize supervised contrastive learning to leverage label information to warm-up the backbone network. In the second stage, we introduce CCFG-Net, a network architecture that integrates classification and contrastive learning in both Euclidean and Angular spaces, in which contrastive learning is applied in both supervised learning and pairwise discrimination manners to enhance the model's feature representation capability. Overall, our proposed approach effectively exploits the complementary strengths of contrastive learning and classification, leading to improved recognition performance on the resembling glyphs. Comparative evaluations with state-of-the-art fine-grained classification approaches under both Convolutional Neural Network (CNN) and Transformer backbones demonstrate the superiority of our proposed method.
Token compression expedites the training and inference of Vision Transformers (ViTs) by reducing the number of the redundant tokens, e.g., pruning inattentive tokens or merging similar tokens. However, when applied to downstream tasks, these approaches suffer from significant performance drop when the compression degrees are mismatched between training and inference stages, which limits the application of token compression on off-the-shelf trained models. In this paper, we propose a model arithmetic framework to decouple the compression degrees between the two stages. In advance, we additionally perform a fast parameter-efficient self-distillation stage on the pre-trained models to obtain a small plugin, called Token Compensator (ToCom), which describes the gap between models across different compression degrees. During inference, ToCom can be directly inserted into any downstream off-the-shelf models with any mismatched training and inference compression degrees to acquire universal performance improvements without further training. Experiments on over 20 downstream tasks demonstrate the effectiveness of our framework. On CIFAR100, fine-grained visual classification, and VTAB-1k, ToCom can yield up to a maximum improvement of 2.3%, 1.5%, and 2.0% in the average performance of DeiT-B, respectively. Code: https://github.com/JieShibo/ToCom
Different from large-scale classification tasks, fine-grained visual classification is a challenging task due to two critical problems: 1) evident intra-class variances and subtle inter-class differences, and 2) overfitting owing to fewer training samples in datasets. Most existing methods extract key features to reduce intra-class variances, but pay no attention to subtle inter-class differences in fine-grained visual classification. To address this issue, we propose a loss function named exploration of class center, which consists of a multiple class-center constraint and a class-center label generation. This loss function fully utilizes the information of the class center from the perspective of features and labels. From the feature perspective, the multiple class-center constraint pulls samples closer to the target class center, and pushes samples away from the most similar nontarget class center. Thus, the constraint reduces intra-class variances and enlarges inter-class differences. From the label perspective, the class-center label generation utilizes classcenter distributions to generate soft labels to alleviate overfitting. Our method can be easily integrated with existing fine-grained visual classification approaches as a loss function, to further boost excellent performance with only slight training costs. Extensive experiments are conducted to demonstrate consistent improvements achieved by our method on four widely-used fine-grained visual classification datasets. In particular, our method achieves state-of-the-art performance on the FGVC-Aircraft and CUB-200-2011 datasets.
Computer vision methods that explicitly detect object parts and reason on them are a step towards inherently interpretable models. Existing approaches that perform part discovery driven by a fine-grained classification task make very restrictive assumptions on the geometric properties of the discovered parts; they should be small and compact. Although this prior is useful in some cases, in this paper we show that pre-trained transformer-based vision models, such as self-supervised DINOv2 ViT, enable the relaxation of these constraints. In particular, we find that a total variation (TV) prior, which allows for multiple connected components of any size, substantially outperforms previous work. We test our approach on three fine-grained classification benchmarks: CUB, PartImageNet and Oxford Flowers, and compare our results to previously published methods as well as a re-implementation of the state-of-the-art method PDiscoNet with a transformer-based backbone. We consistently obtain substantial improvements across the board, both on part discovery metrics and the downstream classification task, showing that the strong inductive biases in self-supervised ViT models require to rethink the geometric priors that can be used for unsupervised part discovery.
The emerging task of fine-grained image classification in low-data regimes assumes the presence of low inter-class variance and large intra-class variation along with a highly limited amount of training samples per class. However, traditional ways of separately dealing with fine-grained categorisation and extremely scarce data may be inefficient under both these harsh conditions presented together. In this paper, we present a novel framework, called AD-Net, aiming to enhance deep neural network performance on this challenge by leveraging the power of Augmentation and Distillation techniques. Specifically, our approach is designed to refine learned features through self-distillation on augmented samples, mitigating harmful overfitting. We conduct comprehensive experiments on popular fine-grained image classification benchmarks where our AD-Net demonstrates consistent improvement over traditional fine-tuning and state-of-the-art low-data techniques. Remarkably, with the smallest data available, our framework shows an outstanding relative accuracy increase of up to 45 % compared to standard ResNet-50 and up to 27 % compared to the closest SOTA runner-up. We emphasise that our approach is practically architecture-independent and adds zero extra cost at inference time. Additionally, we provide an extensive study on the impact of every framework's component, highlighting the importance of each in achieving optimal performance. Source code and trained models are publicly available at github.com/demidovd98/fgic_lowd.
Fine-grained visual classification (FGVC) involves classifying closely related sub-classes. This task is difficult due to the subtle differences between classes and the high intra-class variance. Moreover, FGVC datasets are typically small and challenging to gather, thus highlighting a significant need for effective data augmentation. Recent advancements in text-to-image diffusion models offer new possibilities for augmenting classification datasets. While these models have been used to generate training data for classification tasks, their effectiveness in full-dataset training of FGVC models remains under-explored. Recent techniques that rely on Text2Image generation or Img2Img methods, often struggle to generate images that accurately represent the class while modifying them to a degree that significantly increases the dataset's diversity. To address these challenges, we present SaSPA: Structure and Subject Preserving Augmentation. Contrary to recent methods, our method does not use real images as guidance, thereby increasing generation flexibility and promoting greater diversity. To ensure accurate class representation, we employ conditioning mechanisms, specifically by conditioning on image edges and subject representation. We conduct extensive experiments and benchmark SaSPA against both traditional and recent generative data augmentation methods. SaSPA consistently outperforms all established baselines across multiple settings, including full dataset training, contextual bias, and few-shot classification. Additionally, our results reveal interesting patterns in using synthetic data for FGVC models; for instance, we find a relationship between the amount of real data used and the optimal proportion of synthetic data. Code is available at https://github.com/EyalMichaeli/SaSPA-Aug.
The denoising model has been proven a powerful generative model but has little exploration of discriminative tasks. Representation learning is important in discriminative tasks, which is defined as "learning representations (or features) of the data that make it easier to extract useful information when building classifiers or other predictors". In this paper, we propose a novel Denoising Model for Representation Learning (DenoiseRep) to improve feature discrimination with joint feature extraction and denoising. DenoiseRep views each embedding layer in a backbone as a denoising layer, processing the cascaded embedding layers as if we are recursively denoise features step-by-step. This unifies the frameworks of feature extraction and denoising, where the former progressively embeds features from low-level to high-level, and the latter recursively denoises features step-by-step. After that, DenoiseRep fuses the parameters of feature extraction and denoising layers, and theoretically demonstrates its equivalence before and after the fusion, thus making feature denoising computation-free. DenoiseRep is a label-free algorithm that incrementally improves features but also complementary to the label if available. Experimental results on various discriminative vision tasks, including re-identification (Market-1501, DukeMTMC-reID, MSMT17, CUHK-03, vehicleID), image classification (ImageNet, UB200, Oxford-Pet, Flowers), object detection (COCO), image segmentation (ADE20K) show stability and impressive improvements. We also validate its effectiveness on the CNN (ResNet) and Transformer (ViT, Swin, Vmamda) architectures.
Understanding Activities of Daily Living (ADLs) is a crucial step for different applications including assistive robots, smart homes, and healthcare. However, to date, few benchmarks and methods have focused on complex ADLs, especially those involving multi-person interactions in home environments. In this paper, we propose a new dataset and benchmark, InteractADL, for understanding complex ADLs that involve interaction between humans (and objects). Furthermore, complex ADLs occurring in home environments comprise a challenging long-tailed distribution due to the rarity of multi-person interactions, and pose fine-grained visual recognition tasks due to the presence of semantically and visually similar classes. To address these issues, we propose a novel method for fine-grained few-shot video classification called Name Tuning that enables greater semantic separability by learning optimal class name vectors. We show that Name Tuning can be combined with existing prompt tuning strategies to learn the entire input text (rather than only learning the prompt or class names) and demonstrate improved performance for few-shot classification on InteractADL and 4 other fine-grained visual classification benchmarks. For transparency and reproducibility, we release our code at https://github.com/zanedurante/vlm_benchmark.
This paper revives Densely Connected Convolutional Networks (DenseNets) and reveals the underrated effectiveness over predominant ResNet-style architectures. We believe DenseNets' potential was overlooked due to untouched training methods and traditional design elements not fully revealing their capabilities. Our pilot study shows dense connections through concatenation are strong, demonstrating that DenseNets can be revitalized to compete with modern architectures. We methodically refine suboptimal components - architectural adjustments, block redesign, and improved training recipes towards widening DenseNets and boosting memory efficiency while keeping concatenation shortcuts. Our models, employing simple architectural elements, ultimately surpass Swin Transformer, ConvNeXt, and DeiT-III - key architectures in the residual learning lineage. Furthermore, our models exhibit near state-of-the-art performance on ImageNet-1K, competing with the very recent models and downstream tasks, ADE20k semantic segmentation, and COCO object detection/instance segmentation. Finally, we provide empirical analyses that uncover the merits of the concatenation over additive shortcuts, steering a renewed preference towards DenseNet-style designs. Our code is available at https://github.com/naver-ai/rdnet.
Continual learning requires a model to adapt to ongoing changes in the data distribution, and often to the set of tasks to be performed. It is rare, however, that the data and task changes are completely unpredictable. Given a description of an overarching goal or data theme, which we call a realm, humans can often guess what concepts are associated with it. We show here that the combination of a large language model and an image generation model can similarly provide useful premonitions as to how a continual learning challenge might develop over time. We use the large language model to generate text descriptions of semantically related classes that might potentially appear in the data stream in future. These descriptions are then rendered using Stable Diffusion to generate new labelled image samples. The resulting synthetic dataset is employed for supervised pre-training, but is discarded prior to commencing continual learning, along with the pre-training classification head. We find that the backbone of our pre-trained networks can learn representations useful for the downstream continual learning problem, thus becoming a valuable input to any existing continual learning method. Although there are complexities arising from the domain gap between real and synthetic images, we show that pre-training models in this manner improves multiple Class Incremenal Learning (CIL) methods on fine-grained image classification benchmarks. Supporting code can be found at https://github.com/cl-premonition/premonition.
Vision-language foundation models have exhibited remarkable success across a multitude of downstream tasks due to their scalability on extensive image-text paired data. However, these models also display significant limitations when applied to downstream tasks, such as fine-grained image classification, as a result of ``decision shortcuts'' that hinder their generalization capabilities. In this work, we find that the CLIP model possesses a rich set of features, encompassing both \textit{desired invariant causal features} and \textit{undesired decision shortcuts}. Moreover, the underperformance of CLIP on downstream tasks originates from its inability to effectively utilize pre-trained features in accordance with specific task requirements. To address this challenge, we propose a simple yet effective method, Spurious Feature Eraser (SEraser), to alleviate the decision shortcuts by erasing the spurious features. Specifically, we introduce a test-time prompt tuning paradigm that optimizes a learnable prompt, thereby compelling the model to exploit invariant features while disregarding decision shortcuts during the inference phase. The proposed method effectively alleviates excessive dependence on potentially misleading spurious information. We conduct comparative analysis of the proposed method against various approaches which validates the significant superiority.
Fine-grained vehicle recognition (FGVR) is an essential fundamental technology for intelligent transportation systems, but very difficult because of its inherent intra-class variation. Most previous FGVR studies only focus on the intra-class variation caused by different shooting angles, positions, etc., while the intra-class variation caused by image noise has received little attention. This paper proposes a progressive multi-task anti-noise learning (PMAL) framework and a progressive multi-task distilling (PMD) framework to solve the intra-class variation problem in FGVR due to image noise. The PMAL framework achieves high recognition accuracy by treating image denoising as an additional task in image recognition and progressively forcing a model to learn noise invariance. The PMD framework transfers the knowledge of the PMAL-trained model into the original backbone network, which produces a model with about the same recognition accuracy as the PMAL-trained model, but without any additional overheads over the original backbone network. Combining the two frameworks, we obtain models that significantly exceed previous state-of-the-art methods in recognition accuracy on two widely-used, standard FGVR datasets, namely Stanford Cars, and CompCars, as well as three additional surveillance image-based vehicle-type classification datasets, namely Beijing Institute of Technology (BIT)-Vehicle, Vehicle Type Image Data 2 (VTID2), and Vehicle Images Dataset for Make Model Recognition (VIDMMR), without any additional overheads over the original backbone networks. The source code is available at https://github.com/Dichao-Liu/Anti-noise_FGVR
The zero-shot performance of existing vision-language models (VLMs) such as CLIP is limited by the availability of large-scale, aligned image and text datasets in specific domains. In this work, we leverage two complementary sources of information -- descriptions of categories generated by large language models (LLMs) and abundant, fine-grained image classification datasets -- to improve the zero-shot classification performance of VLMs across fine-grained domains. On the technical side, we develop methods to train VLMs with this "bag-level" image-text supervision. We find that simply using these attributes at test-time does not improve performance, but our training strategy, for example, on the iNaturalist dataset, leads to an average improvement of 4-5% in zero-shot classification accuracy for novel categories of birds and flowers. Similar improvements are observed in domains where a subset of the categories was used to fine-tune the model. By prompting LLMs in various ways, we generate descriptions that capture visual appearance, habitat, and geographic regions and pair them with existing attributes such as the taxonomic structure of the categories. We systematically evaluate their ability to improve zero-shot categorization in natural domains. Our findings suggest that geographic priors can be just as effective and are complementary to visual appearance. Our method also outperforms prior work on prompt-based tuning of VLMs. We release the benchmark, consisting of 14 datasets at https://github.com/cvl-umass/AdaptCLIPZS , which will contribute to future research in zero-shot recognition.