Early identification of weeds is essential for effective management and control, and there is growing interest in automating the process using computer vision techniques coupled with AI methods. However, challenges associated with training AI-based weed identification models, such as limited expert-verified data and complexity and variability in morphological features, have hindered progress. To address these issues, we present WeedNet, the first global-scale weed identification model capable of recognizing an extensive set of weed species, including noxious and invasive plant species. WeedNet is an end-to-end real-time weed identification pipeline and uses self-supervised learning, fine-tuning, and enhanced trustworthiness strategies. WeedNet achieved 91.02% accuracy across 1,593 weed species, with 41% species achieving 100% accuracy. Using a fine-tuning strategy and a Global-to-Local approach, the local Iowa WeedNet model achieved an overall accuracy of 97.38% for 85 Iowa weeds, most classes exceeded a 90% mean accuracy per class. Testing across intra-species dissimilarity (developmental stages) and inter-species similarity (look-alike species) suggests that diversity in the images collected, spanning all the growth stages and distinguishable plant characteristics, is crucial in driving model performance. The generalizability and adaptability of the Global WeedNet model enable it to function as a foundational model, with the Global-to-Local strategy allowing fine-tuning for region-specific weed communities. Additional validation of drone- and ground-rover-based images highlights the potential of WeedNet for integration into robotic platforms. Furthermore, integration with AI for conversational use provides intelligent agricultural and ecological conservation consulting tools for farmers, agronomists, researchers, land managers, and government agencies across diverse landscapes.
Self-supervised learning (SSL) has reduced the reliance on expensive labeling in speech technologies by learning meaningful representations from unannotated data. Since most SSL-based downstream tasks prioritize content information in speech, ideal representations should disentangle content from unwanted variations like speaker characteristics in the SSL representations. However, removing speaker information often degrades other speech components, and existing methods either fail to fully disentangle speaker identity or require resource-intensive models. In this paper, we propose a novel disentanglement method that linearly decomposes SSL representations into speaker-specific and speaker-independent components, effectively generating speaker disentangled representations. Comprehensive experiments show that our approach achieves speaker independence and as such, when applied to content-driven tasks such as voice conversion, our representations yield significant improvements over state-of-the-art methods.
The sustainability of the ocean ecosystem is threatened by increased levels of sound pollution, making monitoring crucial to understand its variability and impact. Passive acoustic monitoring (PAM) systems collect a large amount of underwater sound recordings, but the large volume of data makes manual analysis impossible, creating the need for automation. Although machine learning offers a potential solution, most underwater acoustic recordings are unlabeled. Self-supervised learning models have demonstrated success in learning from large-scale unlabeled data in various domains like computer vision, Natural Language Processing, and audio. However, these models require large, diverse, and balanced datasets for training in order to generalize well. To address this, a fully automated self-supervised data curation pipeline is proposed to create a diverse and balanced dataset from raw PAM data. It integrates Automatic Identification System (AIS) data with recordings from various hydrophones in the U.S. waters. Using hierarchical k-means clustering, the raw audio data is sampled and then combined with AIS samples to create a balanced and diverse dataset. The resulting curated dataset enables the development of self-supervised learning models, facilitating various tasks such as monitoring marine mammals and assessing sound pollution.
Tactile sensing remains far less understood in neuroscience and less effective in artificial systems compared to more mature modalities such as vision and language. We bridge these gaps by introducing a novel Encoder-Attender-Decoder (EAD) framework to systematically explore the space of task-optimized temporal neural networks trained on realistic tactile input sequences from a customized rodent whisker-array simulator. We identify convolutional recurrent neural networks (ConvRNNs) as superior encoders to purely feedforward and state-space architectures for tactile categorization. Crucially, these ConvRNN-encoder-based EAD models achieve neural representations closely matching rodent somatosensory cortex, saturating the explainable neural variability and revealing a clear linear relationship between supervised categorization performance and neural alignment. Furthermore, contrastive self-supervised ConvRNN-encoder-based EADs, trained with tactile-specific augmentations, match supervised neural fits, serving as an ethologically-relevant, label-free proxy. For neuroscience, our findings highlight nonlinear recurrent processing as important for general-purpose tactile representations in somatosensory cortex, providing the first quantitative characterization of the underlying inductive biases in this system. For embodied AI, our results emphasize the importance of recurrent EAD architectures to handle realistic tactile inputs, along with tailored self-supervised learning methods for achieving robust tactile perception with the same type of sensors animals use to sense in unstructured environments.
Chromosome analysis is vital for diagnosing genetic disorders and guiding cancer therapy decisions through the identification of somatic clonal aberrations. However, developing an AI model are hindered by the overwhelming complexity and diversity of chromosomal abnormalities, requiring extensive annotation efforts, while automated methods remain task-specific and lack generalizability due to the scarcity of comprehensive datasets spanning diverse resource conditions. Here, we introduce CHROMA, a foundation model for cytogenomics, designed to overcome these challenges by learning generalizable representations of chromosomal abnormalities. Pre-trained on over 84,000 specimens (~4 million chromosomal images) via self-supervised learning, CHROMA outperforms other methods across all types of abnormalities, even when trained on fewer labelled data and more imbalanced datasets. By facilitating comprehensive mapping of instability and clonal leisons across various aberration types, CHROMA offers a scalable and generalizable solution for reliable and automated clinical analysis, reducing the annotation workload for experts and advancing precision oncology through the early detection of rare genomic abnormalities, enabling broad clinical AI applications and making advanced genomic analysis more accessible.
Deep learning has transformed computer vision but relies heavily on large labeled datasets and computational resources. Transfer learning, particularly fine-tuning pretrained models, offers a practical alternative; however, models pretrained on natural image datasets such as ImageNet may fail to capture domain-specific characteristics in medical imaging. This study introduces an unsupervised learning framework that extracts high-value dermatological features instead of relying solely on ImageNet-based pretraining. We employ a Variational Autoencoder (VAE) trained from scratch on a proprietary dermatological dataset, allowing the model to learn a structured and clinically relevant latent space. This self-supervised feature extractor is then compared to an ImageNet-pretrained backbone under identical classification conditions, highlighting the trade-offs between general-purpose and domain-specific pretraining. Our results reveal distinct learning patterns. The self-supervised model achieves a final validation loss of 0.110 (-33.33%), while the ImageNet-pretrained model stagnates at 0.100 (-16.67%), indicating overfitting. Accuracy trends confirm this: the self-supervised model improves from 45% to 65% (+44.44%) with a near-zero overfitting gap, whereas the ImageNet-pretrained model reaches 87% (+50.00%) but plateaus at 75% (+19.05%), with its overfitting gap increasing to +0.060. These findings suggest that while ImageNet pretraining accelerates convergence, it also amplifies overfitting on non-clinically relevant features. In contrast, self-supervised learning achieves steady improvements, stronger generalization, and superior adaptability, underscoring the importance of domain-specific feature extraction in medical imaging.
The learning mechanisms by which humans acquire internal representations of objects are not fully understood. Deep neural networks (DNNs) have emerged as a useful tool for investigating this question, as they have internal representations similar to those of humans as a byproduct of optimizing their objective functions. While previous studies have shown that models trained with various learning paradigms - such as supervised, self-supervised, and CLIP - acquire human-like representations, it remains unclear whether their similarity to human representations is primarily at a coarse category level or extends to finer details. Here, we employ an unsupervised alignment method based on Gromov-Wasserstein Optimal Transport to compare human and model object representations at both fine-grained and coarse-grained levels. The unique feature of this method compared to conventional representational similarity analysis is that it estimates optimal fine-grained mappings between the representation of each object in human and model representations. We used this unsupervised alignment method to assess the extent to which the representation of each object in humans is correctly mapped to the corresponding representation of the same object in models. Using human similarity judgments of 1,854 objects from the THINGS dataset, we find that models trained with CLIP consistently achieve strong fine- and coarse-grained matching with human object representations. In contrast, self-supervised models showed limited matching at both fine- and coarse-grained levels, but still formed object clusters that reflected human coarse category structure. Our results offer new insights into the role of linguistic information in acquiring precise object representations and the potential of self-supervised learning to capture coarse categorical structures.
In this paper, we focus on the out-of-distribution (OOD) generalization of self-supervised learning (SSL). By analyzing the mini-batch construction during the SSL training phase, we first give one plausible explanation for SSL having OOD generalization. Then, from the perspective of data generation and causal inference, we analyze and conclude that SSL learns spurious correlations during the training process, which leads to a reduction in OOD generalization. To address this issue, we propose a post-intervention distribution (PID) grounded in the Structural Causal Model. PID offers a scenario where the spurious variable and label variable is mutually independent. Besides, we demonstrate that if each mini-batch during SSL training satisfies PID, the resulting SSL model can achieve optimal worst-case OOD performance. This motivates us to develop a batch sampling strategy that enforces PID constraints through the learning of a latent variable model. Through theoretical analysis, we demonstrate the identifiability of the latent variable model and validate the effectiveness of the proposed sampling strategy. Experiments conducted on various downstream OOD tasks demonstrate the effectiveness of the proposed sampling strategy.
Recently, pre-trained models for music information retrieval based on self-supervised learning (SSL) are becoming popular, showing success in various downstream tasks. However, there is limited research on the specific meanings of the encoded information and their applicability. Exploring these aspects can help us better understand their capabilities and limitations, leading to more effective use in downstream tasks. In this study, we analyze the advanced music representation model MusicFM and the newly emerged SSL model MuQ. We focus on three main aspects: (i) validating the advantages of SSL models across multiple downstream tasks, (ii) exploring the specialization of layer-wise information for different tasks, and (iii) comparing performance differences when selecting specific layers. Through this analysis, we reveal insights into the structure and potential applications of SSL models in music information retrieval.
Lung cancer remains among the deadliest types of cancer in recent decades, and early lung nodule detection is crucial for improving patient outcomes. The limited availability of annotated medical imaging data remains a bottleneck in developing accurate computer-aided diagnosis (CAD) systems. Self-supervised learning can help leverage large amounts of unlabeled data to develop more robust CAD systems. With the recent advent of transformer-based architecture and their ability to generalize to unseen tasks, there has been an effort within the healthcare community to adapt them to various medical downstream tasks. Thus, we propose a novel "LungNodule-SSM" method, which utilizes selfsupervised learning with DINOv2 as a backbone to enhance lung nodule detection and classification without annotated data. Our methodology has two stages: firstly, the DINOv2 model is pre-trained on unlabeled CT scans to learn robust feature representations, then secondly, these features are fine-tuned using transformer-based architectures for lesionlevel detection and accurate lung nodule diagnosis. The proposed method has been evaluated on the challenging LUNA 16 dataset, consisting of 888 CT scans, and compared with SOTA methods. Our experimental results show the superiority of our proposed method with an accuracy of 98.37%, explaining its effectiveness in lung nodule detection. The source code, datasets, and pre-processed data can be accessed using the link:https://github.com/EMeRALDsNRPU/Lung-Nodule-SSM-Self-Supervised-Lung-Nodule-Detection-and-Classification/tree/main
Self-supervised learning has emerged as a powerful paradigm for training deep neural networks, particularly in medical imaging where labeled data is scarce. While current approaches typically rely on synthetic augmentations of single images, we propose VET-DINO, a framework that leverages a unique characteristic of medical imaging: the availability of multiple standardized views from the same study. Using a series of clinical veterinary radiographs from the same patient study, we enable models to learn view-invariant anatomical structures and develop an implied 3D understanding from 2D projections. We demonstrate our approach on a dataset of 5 million veterinary radiographs from 668,000 canine studies. Through extensive experimentation, including view synthesis and downstream task performance, we show that learning from real multi-view pairs leads to superior anatomical understanding compared to purely synthetic augmentations. VET-DINO achieves state-of-the-art performance on various veterinary imaging tasks. Our work establishes a new paradigm for self-supervised learning in medical imaging that leverages domain-specific properties rather than merely adapting natural image techniques.
Scene text detection has seen the emergence of high-performing methods that excel on academic benchmarks. However, these detectors often fail to replicate such success in real-world scenarios. We uncover two key factors contributing to this discrepancy through extensive experiments. First, a \textit{Fine-tuning Gap}, where models leverage \textit{Dataset-Specific Optimization} (DSO) paradigm for one domain at the cost of reduced effectiveness in others, leads to inflated performances on academic benchmarks. Second, the suboptimal performance in practical settings is primarily attributed to the long-tailed distribution of texts, where detectors struggle with rare and complex categories as artistic or overlapped text. Given that the DSO paradigm might undermine the generalization ability of models, we advocate for a \textit{Joint-Dataset Learning} (JDL) protocol to alleviate the Fine-tuning Gap. Additionally, an error analysis is conducted to identify three major categories and 13 subcategories of challenges in long-tailed scene text, upon which we propose a Long-Tailed Benchmark (LTB). LTB facilitates a comprehensive evaluation of ability to handle a diverse range of long-tailed challenges. We further introduce MAEDet, a self-supervised learning-based method, as a strong baseline for LTB. The code is available at https://github.com/pd162/LTB.
To overcome the constraints of the underwater environment and improve the accuracy and robustness of underwater target detection models, this paper develops a specialized dataset for underwater target detection and proposes an efficient algorithm for underwater multi-target detection. A self-supervised learning based on the SimSiam structure is employed for the pre-training of underwater target detection network. To address the problems of low detection accuracy caused by low contrast, mutual occlusion and dense distribution of underwater targets in underwater object detection, a detection model suitable for underwater target detection is proposed by introducing deformable convolution and dilated convolution. The proposed detection model can obtain more effective information by increasing the receptive field. In addition, the regression loss function EIoU is introduced, which improves model performance by separately calculating the width and height losses of the predicted box. Experiment results show that the accuracy of the underwater target detection has been improved by the proposed detector.
Recent efforts at scaling computer vision models have established Vision Transformers (ViTs) as the leading architecture. ViTs incorporate weight sharing over image patches as an important inductive bias. In this work, we show that ViTs benefit from incorporating equivariance under the octic group, i.e., reflections and 90-degree rotations, as a further inductive bias. We develop new architectures, octic ViTs, that use octic-equivariant layers and put them to the test on both supervised and self-supervised learning. Through extensive experiments on DeiT-III and DINOv2 training on ImageNet-1K, we show that octic ViTs yield more computationally efficient networks while also improving performance. In particular, we achieve approximately 40% reduction in FLOPs for ViT-H while simultaneously improving both classification and segmentation results.
Self-supervised representation learning for point cloud videos remains a challenging problem with two key limitations: (1) existing methods rely on explicit knowledge to learn motion, resulting in suboptimal representations; (2) prior Masked AutoEncoder (MAE) frameworks struggle to bridge the gap between low-level geometry and high-level dynamics in 4D data. In this work, we propose a novel self-disentangled MAE for learning expressive, discriminative, and transferable 4D representations. To overcome the first limitation, we learn motion by aligning high-level semantics in the latent space \textit{without any explicit knowledge}. To tackle the second, we introduce a \textit{self-disentangled learning} strategy that incorporates the latent token with the geometry token within a shared decoder, effectively disentangling low-level geometry and high-level semantics. In addition to the reconstruction objective, we employ three alignment objectives to enhance temporal understanding, including frame-level motion and video-level global information. We show that our pre-trained encoder surprisingly discriminates spatio-temporal representation without further fine-tuning. Extensive experiments on MSR-Action3D, NTU-RGBD, HOI4D, NvGesture, and SHREC'17 demonstrate the superiority of our approach in both coarse-grained and fine-grained 4D downstream tasks. Notably, Uni4D improves action segmentation accuracy on HOI4D by $+3.8\%$.
Precise classification of megakaryocytes is crucial for diagnosing myelodysplastic syndromes. Although self-supervised learning has shown promise in medical image analysis, its application to classifying megakaryocytes in stained slides faces three main challenges: (1) pervasive background noise that obscures cellular details, (2) a long-tailed distribution that limits data for rare subtypes, and (3) complex morphological variations leading to high intra-class variability. To address these issues, we propose the ActiveSSF framework, which integrates active learning with self-supervised pretraining. Specifically, our approach employs Gaussian filtering combined with K-means clustering and HSV analysis (augmented by clinical prior knowledge) for accurate region-of-interest extraction; an adaptive sample selection mechanism that dynamically adjusts similarity thresholds to mitigate class imbalance; and prototype clustering on labeled samples to overcome morphological complexity. Experimental results on clinical megakaryocyte datasets demonstrate that ActiveSSF not only achieves state-of-the-art performance but also significantly improves recognition accuracy for rare subtypes. Moreover, the integration of these advanced techniques further underscores the practical potential of ActiveSSF in clinical settings.
Supervised learning demands large amounts of precisely annotated data to achieve promising results. Such data curation is labor-intensive and imposes significant overhead regarding time and costs. Self-supervised learning (SSL) partially overcomes these limitations by exploiting vast amounts of unlabeled data and creating surrogate (pretext or proxy) tasks to learn useful representations without manual labeling. As a result, SSL has become a powerful machine learning (ML) paradigm for solving several practical downstream computer vision problems, such as classification, detection, and segmentation. Image segmentation is the cornerstone of many high-level visual perception applications, including medical imaging, intelligent transportation, agriculture, and surveillance. Although there is substantial research potential for developing advanced algorithms for SSL-based semantic segmentation, a comprehensive study of existing methodologies is essential to trace advances and guide emerging researchers. This survey thoroughly investigates over 150 recent image segmentation articles, particularly focusing on SSL. It provides a practical categorization of pretext tasks, downstream tasks, and commonly used benchmark datasets for image segmentation research. It concludes with key observations distilled from a large body of literature and offers future directions to make this research field more accessible and comprehensible for readers.
We introduce Perceptual-Initialization (PI), a paradigm shift in visual representation learning that incorporates human perceptual structure during the initialization phase rather than as a downstream fine-tuning step. By integrating human-derived triplet embeddings from the NIGHTS dataset to initialize a CLIP vision encoder, followed by self-supervised learning on YFCC15M, our approach demonstrates significant zero-shot performance improvements, without any task-specific fine-tuning, across 29 zero shot classification and 2 retrieval benchmarks. On ImageNet-1K, zero-shot gains emerge after approximately 15 epochs of pretraining. Benefits are observed across datasets of various scales, with improvements manifesting at different stages of the pretraining process depending on dataset characteristics. Our approach consistently enhances zero-shot top-1 accuracy, top-5 accuracy, and retrieval recall (e.g., R@1, R@5) across these diverse evaluation tasks, without requiring any adaptation to target domains. These findings challenge the conventional wisdom of using human-perceptual data primarily for fine-tuning and demonstrate that embedding human perceptual structure during early representation learning yields more capable and vision-language aligned systems that generalize immediately to unseen tasks. Our work shows that "beginning with you", starting with human perception, provides a stronger foundation for general-purpose vision-language intelligence.
Compared to light-field microscopy (LFM), which enables high-speed volumetric imaging but suffers from non-uniform spatial sampling, Fourier light-field microscopy (FLFM) introduces sub-aperture division at the pupil plane, thereby ensuring spatially invariant sampling and enhancing spatial resolution. Conventional FLFM reconstruction methods, such as Richardson-Lucy (RL) deconvolution, exhibit poor axial resolution and signal degradation due to the ill-posed nature of the inverse problem. While data-driven approaches enhance spatial resolution by leveraging high-quality paired datasets or imposing structural priors, Neural Radiance Fields (NeRF)-based methods employ physics-informed self-supervised learning to overcome these limitations, yet they are hindered by substantial computational costs and memory demands. Therefore, we propose 3D Gaussian Adaptive Tomography (3DGAT) for FLFM, a 3D gaussian splatting based self-supervised learning framework that significantly improves the volumetric reconstruction quality of FLFM while maintaining computational efficiency. Experimental results indicate that our approach achieves higher resolution and improved reconstruction accuracy, highlighting its potential to advance FLFM imaging and broaden its applications in 3D optical microscopy.
This research confronts the challenge of substantial physical equation discrepancies encountered in the generation of spatiotemporal physical fields through data-driven trained models. A spatiotemporal physical field generation model, named HMT-PF, is developed based on the hybrid Mamba-Transformer architecture, incorporating unstructured grid information as input. A fine-tuning block, enhanced with physical information, is introduced to effectively reduce the physical equation discrepancies. The physical equation residuals are computed through a point query mechanism for efficient gradient evaluation, then encoded into latent space for refinement. The fine-tuning process employs a self-supervised learning approach to achieve physical consistency while maintaining essential field characteristics. Results show that the hybrid Mamba-Transformer model achieves good performance in generating spatiotemporal fields, while the physics-informed fine-tuning mechanism further reduces significant physical errors effectively. A MSE-R evaluation method is developed to assess the accuracy and realism of physical field generation.