Muhammad Zeshan Afzal

Tahira Shehzadi, Khurram Azeem Hashmi, Didier Stricker et al.

The astounding performance of transformers in natural language processing (NLP) has motivated researchers to explore their applications in computer vision tasks. DEtection TRansformer (DETR) introduces transformers to object detection tasks by reframing detection as a set prediction problem. Consequently, eliminating the need for proposal generation and post-processing steps. Initially, despite competitive performance, DETR suffered from slow training convergence and ineffective detection of smaller objects. However, numerous improvements are proposed to address these issues, leading to substantial improvements in DETR and enabling it to exhibit state-of-the-art performance. To our knowledge, this is the first paper to provide a comprehensive review of 21 recently proposed advancements in the original DETR model. We dive into both the foundational modules of DETR and its recent enhancements, such as modifications to the backbone structure, query design strategies, and refinements to attention mechanisms. Moreover, we conduct a comparative analysis across various detection transformers, evaluating their performance and network architectures. We hope that this study will ignite further interest among researchers in addressing the existing challenges and exploring the application of transformers in the object detection domain. Readers interested in the ongoing developments in detection transformers can refer to our website at: https://github.com/mindgarage-shan/trans_object_detection_survey

Muhammad Ferjad Naeem, Muhammad Gul Zain Ali Khan, Yongqin Xian et al.

Recent works have shown that unstructured text (documents) from online sources can serve as useful auxiliary information for zero-shot image classification. However, these methods require access to a high-quality source like Wikipedia and are limited to a single source of information. Large Language Models (LLM) trained on web-scale text show impressive abilities to repurpose their learned knowledge for a multitude of tasks. In this work, we provide a novel perspective on using an LLM to provide text supervision for a zero-shot image classification model. The LLM is provided with a few text descriptions from different annotators as examples. The LLM is conditioned on these examples to generate multiple text descriptions for each class(referred to as views). Our proposed model, I2MVFormer, learns multi-view semantic embeddings for zero-shot image classification with these class views. We show that each text view of a class provides complementary information allowing a model to learn a highly discriminative class embedding. Moreover, we show that I2MVFormer is better at consuming the multi-view text supervision from LLM compared to baseline models. I2MVFormer establishes a new state-of-the-art on three public benchmark datasets for zero-shot image classification with unsupervised semantic embeddings.

Muhammad Gul Zain Ali Khan, Muhammad Ferjad Naeem, Luc Van Gool et al.

Continual Learning aims to learn a single model on a sequence of tasks without having access to data from previous tasks. The biggest challenge in the domain still remains catastrophic forgetting: a loss in performance on seen classes of earlier tasks. Some existing methods rely on an expensive replay buffer to store a chunk of data from previous tasks. This, while promising, becomes expensive when the number of tasks becomes large or data can not be stored for privacy reasons. As an alternative, prompt-based methods have been proposed that store the task information in a learnable prompt pool. This prompt pool instructs a frozen image encoder on how to solve each task. While the model faces a disjoint set of classes in each task in this setting, we argue that these classes can be encoded to the same embedding space of a pre-trained language encoder. In this work, we propose Language Guidance for Prompt-based Continual Learning (LGCL) as a plug-in for prompt-based methods. LGCL is model agnostic and introduces language guidance at the task level in the prompt pool and at the class level on the output feature of the vision encoder. We show with extensive experimentation that LGCL consistently improves the performance of prompt-based continual learning methods to set a new state-of-the art. LGCL achieves these performance improvements without needing any additional learnable parameters.

Muhammad Gul Zain Ali Khan, Muhammad Ferjad Naeem, Luc Van Gool et al.

Compositional zero-shot learning aims to recognize unseen compositions of seen visual primitives of object classes and their states. While all primitives (states and objects) are observable during training in some combination, their complex interaction makes this task especially hard. For example, wet changes the visual appearance of a dog very differently from a bicycle. Furthermore, we argue that relationships between compositions go beyond shared states or objects. A cluttered office can contain a busy table; even though these compositions don't share a state or object, the presence of a busy table can guide the presence of a cluttered office. We propose a novel method called Compositional Attention Propagated Embedding (CAPE) as a solution. The key intuition to our method is that a rich dependency structure exists between compositions arising from complex interactions of primitives in addition to other dependencies between compositions. CAPE learns to identify this structure and propagates knowledge between them to learn class embedding for all seen and unseen compositions. In the challenging generalized compositional zero-shot setting, we show that our method outperforms previous baselines to set a new state-of-the-art on three publicly available benchmarks.

Muhammad Usama, Didier Stricker, Mohammad Sadil Khan et al.

Learning representations of CAD models is a largely open problem. While 3D representation learning has flourished around point clouds and meshes, the native format of CAD - boundary representations BReps, which encodes exact parametric surfaces, curves, and their topology, has received little attention as a representation learning substrate. We introduce BRepCLIP, the first framework to align BRep geometry with language and image embeddings through contrastive pretraining. We model each CAD object as a sequence of face and edge tokens with separate discrete vocabularies for surface and curve geometry, augmented with spatial and semantic descriptors that capture surface types (e.g., cylindrical, torus, NURBS) and curve primitives (e.g., line, arc, B-spline). A transformer encoder aggregates these tokens into a global BRep embedding, aligned with CLIP's text and image encoders via a joint contrastive objective. BRepCLIP generates more discriminative and semantically grounded embeddings than existing point-based alternatives, improving Top-1 retrieval over OpenShape by 40.4%, 22.0%, and 23.9% on ABC, CADParser, and Automate, respectively, and improving zero-shot classification on FabWave by 15% in Top-1 score. We further demonstrate its utility as a CAD-aware similarity metric for evaluating text and image-conditioned CAD generation, establishing the importance of structure-aware pretraining for multimodal CAD understanding. Project page is available at https://muhammadusama100.github.io/BrepClip2026/

Danish Nazir, Marcus Liwicki, Didier Stricker et al.

Depth completion involves recovering a dense depth map from a sparse map and an RGB image. Recent approaches focus on utilizing color images as guidance images to recover depth at invalid pixels. However, color images alone are not enough to provide the necessary semantic understanding of the scene. Consequently, the depth completion task suffers from sudden illumination changes in RGB images (e.g., shadows). In this paper, we propose a novel three-branch backbone comprising color-guided, semantic-guided, and depth-guided branches. Specifically, the color-guided branch takes a sparse depth map and RGB image as an input and generates color depth which includes color cues (e.g., object boundaries) of the scene. The predicted dense depth map of color-guided branch along-with semantic image and sparse depth map is passed as input to semantic-guided branch for estimating semantic depth. The depth-guided branch takes sparse, color, and semantic depths to generate the dense depth map. The color depth, semantic depth, and guided depth are adaptively fused to produce the output of our proposed three-branch backbone. In addition, we also propose to apply semantic-aware multi-modal attention-based fusion block (SAMMAFB) to fuse features between all three branches. We further use CSPN++ with Atrous convolutions to refine the dense depth map produced by our three-branch backbone. Extensive experiments show that our model achieves state-of-the-art performance in the KITTI depth completion benchmark at the time of submission.

Tahira Shehzadi, Khurram Azeem Hashmi, Didier Stricker et al.

This paper takes an important step in bridging the performance gap between DETR and R-CNN for graphical object detection. Existing graphical object detection approaches have enjoyed recent enhancements in CNN-based object detection methods, achieving remarkable progress. Recently, Transformer-based detectors have considerably boosted the generic object detection performance, eliminating the need for hand-crafted features or post-processing steps such as Non-Maximum Suppression (NMS) using object queries. However, the effectiveness of such enhanced transformer-based detection algorithms has yet to be verified for the problem of graphical object detection. Essentially, inspired by the latest advancements in the DETR, we employ the existing detection transformer with few modifications for graphical object detection. We modify object queries in different ways, using points, anchor boxes and adding positive and negative noise to the anchors to boost performance. These modifications allow for better handling of objects with varying sizes and aspect ratios, more robustness to small variations in object positions and sizes, and improved image discrimination between objects and non-objects. We evaluate our approach on the four graphical datasets: PubTables, TableBank, NTable and PubLaynet. Upon integrating query modifications in the DETR, we outperform prior works and achieve new state-of-the-art results with the mAP of 96.9\%, 95.7\% and 99.3\% on TableBank, PubLaynet, PubTables, respectively. The results from extensive ablations show that transformer-based methods are more effective for document analysis analogous to other applications. We hope this study draws more attention to the research of using detection transformers in document image analysis.

Mohammad Sadil Khan, Sankalp Sinha, Talha Uddin Sheikh et al.

Prototyping complex computer-aided design (CAD) models in modern softwares can be very time-consuming. This is due to the lack of intelligent systems that can quickly generate simpler intermediate parts. We propose Text2CAD, the first AI framework for generating text-to-parametric CAD models using designer-friendly instructions for all skill levels. Furthermore, we introduce a data annotation pipeline for generating text prompts based on natural language instructions for the DeepCAD dataset using Mistral and LLaVA-NeXT. The dataset contains $\sim170$K models and $\sim660$K text annotations, from abstract CAD descriptions (e.g., generate two concentric cylinders) to detailed specifications (e.g., draw two circles with center $(x,y)$ and radius $r_{1}$, $r_{2}$, and extrude along the normal by $d$...). Within the Text2CAD framework, we propose an end-to-end transformer-based auto-regressive network to generate parametric CAD models from input texts. We evaluate the performance of our model through a mixture of metrics, including visual quality, parametric precision, and geometrical accuracy. Our proposed framework shows great potential in AI-aided design applications. Our source code and annotations will be publicly available.

Tahira Shehzadi, Ifza, Didier Stricker et al.

The impressive advancements in semi-supervised learning have driven researchers to explore its potential in object detection tasks within the field of computer vision. Semi-Supervised Object Detection (SSOD) leverages a combination of a small labeled dataset and a larger, unlabeled dataset. This approach effectively reduces the dependence on large labeled datasets, which are often expensive and time-consuming to obtain. Initially, SSOD models encountered challenges in effectively leveraging unlabeled data and managing noise in generated pseudo-labels for unlabeled data. However, numerous recent advancements have addressed these issues, resulting in substantial improvements in SSOD performance. This paper presents a comprehensive review of 27 cutting-edge developments in SSOD methodologies, from Convolutional Neural Networks (CNNs) to Transformers. We delve into the core components of semi-supervised learning and its integration into object detection frameworks, covering data augmentation techniques, pseudo-labeling strategies, consistency regularization, and adversarial training methods. Furthermore, we conduct a comparative analysis of various SSOD models, evaluating their performance and architectural differences. We aim to ignite further research interest in overcoming existing challenges and exploring new directions in semi-supervised learning for object detection.

Shalini Sarode, Muhammad Saif Ullah Khan, Tahira Shehzadi et al.

We propose ClassroomKD, a novel multi-mentor knowledge distillation framework inspired by classroom environments to enhance knowledge transfer between the student and multiple mentors with different knowledge levels. Unlike traditional methods that rely on fixed mentor-student relationships, our framework dynamically selects and adapts the teaching strategies of diverse mentors based on their effectiveness for each data sample. ClassroomKD comprises two main modules: the Knowledge Filtering (KF) module and the Mentoring module. The KF Module dynamically ranks mentors based on their performance for each input, activating only high-quality mentors to minimize error accumulation and prevent information loss. The Mentoring Module adjusts the distillation strategy by tuning each mentor's influence according to the dynamic performance gap between the student and mentors, effectively modulating the learning pace. Extensive experiments on image classification (CIFAR-100 and ImageNet) and 2D human pose estimation (COCO Keypoints and MPII Human Pose) demonstrate that ClassroomKD outperforms existing knowledge distillation methods for different network architectures. Our results highlight that a dynamic and adaptive approach to mentor selection and guidance leads to more effective knowledge transfer, paving the way for enhanced model performance through distillation.

Muhammad Ahmed Ullah Khan, Muhammad Haris Bin Amir, Didier Stricker et al.

Continual learning enables models to acquire new knowledge over time while retaining previously learned capabilities. However, its application to text-to-3D generation remains unexplored. We present ReConText3D, the first framework for continual text-to-3D generation. We first demonstrate that existing text-to-3D models suffer from catastrophic forgetting under incremental training. ReConText3D enables generative models to incrementally learn new 3D categories from textual descriptions while preserving the ability to synthesize previously seen assets. Our method constructs a compact and diverse replay memory through text-embedding k-Center selection, allowing representative rehearsal of prior knowledge without modifying the underlying architecture. To systematically evaluate continual text-to-3D learning, we introduce Toys4K-CL, a benchmark derived from the Toys4K dataset that provides balanced and semantically diverse class-incremental splits. Extensive experiments on the Toys4K-CL benchmark show that ReConText3D consistently outperforms all baselines across different generative backbones, maintaining high-quality generation for both old and new classes. To the best of our knowledge, this work establishes the first continual learning framework and benchmark for text-to-3D generation, opening a new direction for incremental 3D generative modeling. Project page is available at: https://mauk95.github.io/ReConText3D/.

Hamam Mokayed, Palaiahnakote Shivakumara, Lama Alkhaled et al.

Vehicle detection in real-time scenarios is challenging because of the time constraints and the presence of multiple types of vehicles with different speeds, shapes, structures, etc. This paper presents a new method relied on generating a confidence map-for robust and faster vehicle detection. To reduce the adverse effect of different speeds, shapes, structures, and the presence of several vehicles in a single image, we introduce the concept of augmentation which highlights the region of interest containing the vehicles. The augmented map is generated by exploring the combination of multiresolution analysis and maximally stable extremal regions (MR-MSER). The output of MR-MSER is supplied to fast CNN to generate a confidence map, which results in candidate regions. Furthermore, unlike existing models that implement complicated models for vehicle detection, we explore the combination of a rough set and fuzzy-based models for robust vehicle detection. To show the effectiveness of the proposed method, we conduct experiments on our dataset captured by drones and on several vehicle detection benchmark datasets, namely, KITTI and UA-DETRAC. The results on our dataset and the benchmark datasets show that the proposed method outperforms the existing methods in terms of time efficiency and achieves a good detection rate.

Muhammad Usama, Mohammad Sadil Khan, Didier Stricker et al.

Generating editable 3D CAD models from natural language remains challenging, as existing text-to-CAD systems either produce meshes or rely on scarce design-history data. We present NURBGen, the first framework to generate high-fidelity 3D CAD models directly from text using Non-Uniform Rational B-Splines (NURBS). To achieve this, we fine-tune a large language model (LLM) to translate free-form texts into JSON representations containing NURBS surface parameters (\textit{i.e}, control points, knot vectors, degrees, and rational weights) which can be directly converted into BRep format using Python. We further propose a hybrid representation that combines untrimmed NURBS with analytic primitives to handle trimmed surfaces and degenerate regions more robustly, while reducing token complexity. Additionally, we introduce partABC, a curated subset of the ABC dataset consisting of individual CAD components, annotated with detailed captions using an automated annotation pipeline. NURBGen demonstrates strong performance on diverse prompts, surpassing prior methods in geometric fidelity and dimensional accuracy, as confirmed by expert evaluations. Code and dataset will be released publicly.

Sankalp Sinha, Mohammad Sadil Khan, Muhammad Usama et al.

Generating high-fidelity 3D content from text prompts remains a significant challenge in computer vision due to the limited size, diversity, and annotation depth of the existing datasets. To address this, we introduce MARVEL-40M+, an extensive dataset with 40 million text annotations for over 8.9 million 3D assets aggregated from seven major 3D datasets. Our contribution is a novel multi-stage annotation pipeline that integrates open-source pretrained multi-view VLMs and LLMs to automatically produce multi-level descriptions, ranging from detailed (150-200 words) to concise semantic tags (10-20 words). This structure supports both fine-grained 3D reconstruction and rapid prototyping. Furthermore, we incorporate human metadata from source datasets into our annotation pipeline to add domain-specific information in our annotation and reduce VLM hallucinations. Additionally, we develop MARVEL-FX3D, a two-stage text-to-3D pipeline. We fine-tune Stable Diffusion with our annotations and use a pretrained image-to-3D network to generate 3D textured meshes within 15s. Extensive evaluations show that MARVEL-40M+ significantly outperforms existing datasets in annotation quality and linguistic diversity, achieving win rates of 72.41% by GPT-4 and 73.40% by human evaluators. Project page is available at https://sankalpsinha-cmos.github.io/MARVEL/.

Zhijing Lu, Khurram Azeem Hashmi, Didier Stricker et al.

Unsupervised pixel-level video understanding remains challenging in real-world scenarios, where motion blur, occlusion, and fast object dynamics often cause temporal drift and flickering pseudo-labels.We propose VVitCutLER, an unsupervised framework for video object detection and instance segmentation, which improves the quality of pseudo-labels through temporal consistency. Our core contribution is VitCut, a temporarily stable pseudo-label generator that reduces error accumulation during field degradation through cross-frame region consistency. Meanwhile, VitCut uses a distillation decoder to achieve effective instance mask prediction. Then, based on VitCut, VVitCutLER further integrates cross-frame feature aggregation to enhance video-level robustness. Extensive experiments on standard video benchmarks demonstrate that VVitCutLER significantly improves detection and segmentation performance while reducing temporal instability. These results highlight the importance of temporally consistent supervision for robust pixel-level video understanding.

Muhammad Saif Ullah Khan, Sankalp Sinha, Didier Stricker et al.

Reconstructing texture-less surfaces poses unique challenges in computer vision, primarily due to the lack of specialized datasets that cater to the nuanced needs of depth and normals estimation in the absence of textural information. We introduce "Shape2.5D," a novel, large-scale dataset designed to address this gap. Comprising 1.17 million frames spanning over 39,772 3D models and 48 unique objects, our dataset provides depth and surface normal maps for texture-less object reconstruction. The proposed dataset includes synthetic images rendered with 3D modeling software to simulate various lighting conditions and viewing angles. It also includes a real-world subset comprising 4,672 frames captured with a depth camera. Our comprehensive benchmarks demonstrate the dataset's ability to support the development of algorithms that robustly estimate depth and normals from RGB images and perform voxel reconstruction. Our open-source data generation pipeline allows the dataset to be extended and adapted for future research. The dataset is publicly available at https://github.com/saifkhichi96/Shape25D.

Sheraz Ahmed, Muhammad Imran Malik, Muhammad Zeshan Afzal et al.

The contribution of this paper is fourfold. The first contribution is a novel, generic method for automatic ground truth generation of camera-captured document images (books, magazines, articles, invoices, etc.). It enables us to build large-scale (i.e., millions of images) labeled camera-captured/scanned documents datasets, without any human intervention. The method is generic, language independent and can be used for generation of labeled documents datasets (both scanned and cameracaptured) in any cursive and non-cursive language, e.g., English, Russian, Arabic, Urdu, etc. To assess the effectiveness of the presented method, two different datasets in English and Russian are generated using the presented method. Evaluation of samples from the two datasets shows that 99:98% of the images were correctly labeled. The second contribution is a large dataset (called C3Wi) of camera-captured characters and words images, comprising 1 million word images (10 million character images), captured in a real camera-based acquisition. This dataset can be used for training as well as testing of character recognition systems on camera-captured documents. The third contribution is a novel method for the recognition of cameracaptured document images. The proposed method is based on Long Short-Term Memory and outperforms the state-of-the-art methods for camera based OCRs. As a fourth contribution, various benchmark tests are performed to uncover the behavior of commercial (ABBYY), open source (Tesseract), and the presented camera-based OCR using the presented C3Wi dataset. Evaluation results reveal that the existing OCRs, which already get very high accuracies on scanned documents, have limited performance on camera-captured document images; where ABBYY has an accuracy of 75%, Tesseract an accuracy of 50.22%, while the presented character recognition system has an accuracy of 95.10%.

Tahira Shehzadi, Khurram Azeem Hashmi, Didier Stricker et al.

In this paper, we address the limitations of the DETR-based semi-supervised object detection (SSOD) framework, particularly focusing on the challenges posed by the quality of object queries. In DETR-based SSOD, the one-to-one assignment strategy provides inaccurate pseudo-labels, while the one-to-many assignments strategy leads to overlapping predictions. These issues compromise training efficiency and degrade model performance, especially in detecting small or occluded objects. We introduce Sparse Semi-DETR, a novel transformer-based, end-to-end semi-supervised object detection solution to overcome these challenges. Sparse Semi-DETR incorporates a Query Refinement Module to enhance the quality of object queries, significantly improving detection capabilities for small and partially obscured objects. Additionally, we integrate a Reliable Pseudo-Label Filtering Module that selectively filters high-quality pseudo-labels, thereby enhancing detection accuracy and consistency. On the MS-COCO and Pascal VOC object detection benchmarks, Sparse Semi-DETR achieves a significant improvement over current state-of-the-art methods that highlight Sparse Semi-DETR's effectiveness in semi-supervised object detection, particularly in challenging scenarios involving small or partially obscured objects.

Tahira Shehzadi, Didier Stricker, Muhammad Zeshan Afzal

Document layout analysis involves understanding the arrangement of elements within a document. This paper navigates the complexities of understanding various elements within document images, such as text, images, tables, and headings. The approach employs an advanced Transformer-based object detection network as an innovative graphical page object detector for identifying tables, figures, and displayed elements. We introduce a query encoding mechanism to provide high-quality object queries for contrastive learning, enhancing efficiency in the decoder phase. We also present a hybrid matching scheme that integrates the decoder's original one-to-one matching strategy with the one-to-many matching strategy during the training phase. This approach aims to improve the model's accuracy and versatility in detecting various graphical elements on a page. Our experiments on PubLayNet, DocLayNet, and PubTables benchmarks show that our approach outperforms current state-of-the-art methods. It achieves an average precision of 97.3% on PubLayNet, 81.6% on DocLayNet, and 98.6 on PubTables, demonstrating its superior performance in layout analysis. These advancements not only enhance the conversion of document images into editable and accessible formats but also streamline information retrieval and data extraction processes.

Iqraa Ehsan, Tahira Shehzadi, Didier Stricker et al.

Table detection, a pivotal task in document analysis, aims to precisely recognize and locate tables within document images. Although deep learning has shown remarkable progress in this realm, it typically requires an extensive dataset of labeled data for proficient training. Current CNN-based semi-supervised table detection approaches use the anchor generation process and Non-Maximum Suppression (NMS) in their detection process, limiting training efficiency. Meanwhile, transformer-based semi-supervised techniques adopted a one-to-one match strategy that provides noisy pseudo-labels, limiting overall efficiency. This study presents an innovative transformer-based semi-supervised table detector. It improves the quality of pseudo-labels through a novel matching strategy combining one-to-one and one-to-many assignment techniques. This approach significantly enhances training efficiency during the early stages, ensuring superior pseudo-labels for further training. Our semi-supervised approach is comprehensively evaluated on benchmark datasets, including PubLayNet, ICADR-19, and TableBank. It achieves new state-of-the-art results, with a mAP of 95.7% and 97.9% on TableBank (word) and PubLaynet with 30% label data, marking a 7.4 and 7.6 point improvement over previous semi-supervised table detection approach, respectively. The results clearly show the superiority of our semi-supervised approach, surpassing all existing state-of-the-art methods by substantial margins. This research represents a significant advancement in semi-supervised table detection methods, offering a more efficient and accurate solution for practical document analysis tasks.

Tahira Shehzadi, Shalini Sarode, Didier Stricker et al.

Table detection within document images is a crucial task in document processing, involving the identification and localization of tables. Recent strides in deep learning have substantially improved the accuracy of this task, but it still heavily relies on large labeled datasets for effective training. Several semi-supervised approaches have emerged to overcome this challenge, often employing CNN-based detectors with anchor proposals and post-processing techniques like non-maximal suppression (NMS). However, recent advancements in the field have shifted the focus towards transformer-based techniques, eliminating the need for NMS and emphasizing object queries and attention mechanisms. Previous research has focused on two key areas to improve transformer-based detectors: refining the quality of object queries and optimizing attention mechanisms. However, increasing object queries can introduce redundancy, while adjustments to the attention mechanism can increase complexity. To address these challenges, we introduce a semi-supervised approach employing SAM-DETR, a novel approach for precise alignment between object queries and target features. Our approach demonstrates remarkable reductions in false positives and substantial enhancements in table detection performance, particularly in complex documents characterized by diverse table structures. This work provides more efficient and accurate table detection in semi-supervised settings.

Muhammad Saif Ullah Khan, Muhammad Ferjad Naeem, Federico Tombari et al.

We present a novel LLM-based pipeline for creating contextual descriptions of human body poses in images using only auxiliary attributes. This approach facilitates the creation of the MPII Pose Descriptions dataset, which includes natural language annotations for 17,367 images containing people engaged in 410 distinct activities. We demonstrate the effectiveness of our pose descriptions in enabling zero-shot human-centric classification using CLIP. Moreover, we introduce the FocusCLIP framework, which incorporates Subject-Focused Attention (SFA) in CLIP for improved text-to-image alignment. Our models were pretrained on the MPII Pose Descriptions dataset and their zero-shot performance was evaluated on five unseen datasets covering three tasks. FocusCLIP outperformed the baseline CLIP model, achieving an average accuracy increase of 8.61\% (33.65\% compared to CLIP's 25.04\%). Notably, our approach yielded improvements of 3.98\% in activity recognition, 14.78\% in age classification, and 7.06\% in emotion recognition. These results highlight the potential of integrating detailed pose descriptions and subject-level guidance into general pretraining frameworks for enhanced performance in downstream tasks.

Khurram Azeem Hashmi, Talha Uddin Sheikh, Didier Stricker et al.

The primary challenge in Video Object Detection (VOD) is effectively exploiting temporal information to enhance object representations. Traditional strategies, such as aggregating region proposals, often suffer from feature variance due to the inclusion of background information. We introduce a novel instance mask-based feature aggregation approach, significantly refining this process and deepening the understanding of object dynamics across video frames. We present FAIM, a new VOD method that enhances temporal Feature Aggregation by leveraging Instance Mask features. In particular, we propose the lightweight Instance Feature Extraction Module (IFEM) to learn instance mask features and the Temporal Instance Classification Aggregation Module (TICAM) to aggregate instance mask and classification features across video frames. Using YOLOX as a base detector, FAIM achieves 87.9% mAP on the ImageNet VID dataset at 33 FPS on a single 2080Ti GPU, setting a new benchmark for the speed-accuracy trade-off. Additional experiments on multiple datasets validate that our approach is robust, method-agnostic, and effective in multi-object tracking, demonstrating its broader applicability to video understanding tasks.

Muhammad Saif Ullah Khan, Tahira Shehzadi, Rabeya Noor et al.

Automated signature verification on bank checks is critical for fraud prevention and ensuring transaction authenticity. This task is challenging due to the coexistence of signatures with other textual and graphical elements on real-world documents. Verification systems must first detect the signature and then validate its authenticity, a dual challenge often overlooked by current datasets and methodologies focusing only on verification. To address this gap, we introduce a novel dataset specifically designed for signature verification on bank checks. This dataset includes a variety of signature styles embedded within typical check elements, providing a realistic testing ground for advanced detection methods. Moreover, we propose a novel approach for writer-independent signature verification using an object detection network. Our detection-based verification method treats genuine and forged signatures as distinct classes within an object detection framework, effectively handling both detection and verification. We employ a DINO-based network augmented with a dilation module to detect and verify signatures on check images simultaneously. Our approach achieves an AP of 99.2 for genuine and 99.4 for forged signatures, a significant improvement over the DINO baseline, which scored 93.1 and 89.3 for genuine and forged signatures, respectively. This improvement highlights our dilation module's effectiveness in reducing both false positives and negatives. Our results demonstrate substantial advancements in detection-based signature verification technology, offering enhanced security and efficiency in financial document processing.

Muhammad Saif Ullah Khan, Muhammad Zeshan Afzal, Didier Stricker

Large language models (LLMs) have achieved remarkable success in text-based tasks but often struggle to provide actionable guidance in real-world physical environments. This is because of their inability to recognize their limited understanding of the user's physical context. We present SituationalLLM, a novel approach that integrates structured scene information into an LLM to deliver proactive, context-aware assistance. By encoding objects, attributes, and relationships in a custom Scene Graph Language, SituationalLLM actively identifies gaps in environmental context and seeks clarifications during user interactions. This behavior emerges from training on the Situational Awareness Database for Instruct-Tuning (SAD-Instruct), which combines diverse, scenario-specific scene graphs with iterative, dialogue-based refinements. Experimental results indicate that SituationalLLM outperforms generic LLM baselines in task specificity, reliability, and adaptability, paving the way for environment-aware AI assistants capable of delivering robust, user-centric guidance under real-world constraints.

Talha Uddin Sheikh, Tahira Shehzadi, Khurram Azeem Hashmi et al.

Document layout analysis is a key area in document research, involving techniques like text mining and visual analysis. Despite various methods developed to tackle layout analysis, a critical but frequently overlooked problem is the scarcity of labeled data needed for analyses. With the rise of internet use, an overwhelming number of documents are now available online, making the process of accurately labeling them for research purposes increasingly challenging and labor-intensive. Moreover, the diversity of documents online presents a unique set of challenges in maintaining the quality and consistency of these labels, further complicating document layout analysis in the digital era. To address this, we employ a vision-based approach for analyzing document layouts designed to train a network without labels. Instead, we focus on pre-training, initially generating simple object masks from the unlabeled document images. These masks are then used to train a detector, enhancing object detection and segmentation performance. The model's effectiveness is further amplified through several unsupervised training iterations, continuously refining its performance. This approach significantly advances document layout analysis, particularly precision and efficiency, without labels.

Sankalp Sinha, Muhammad Saif Ullah Khan, Talha Uddin Sheikh et al.

Zero-shot learning has been extensively investigated in the broader field of visual recognition, attracting significant interest recently. However, the current work on zero-shot learning in document image classification remains scarce. The existing studies either focus exclusively on zero-shot inference, or their evaluation does not align with the established criteria of zero-shot evaluation in the visual recognition domain. We provide a comprehensive document image classification analysis in Zero-Shot Learning (ZSL) and Generalized Zero-Shot Learning (GZSL) settings to address this gap. Our methodology and evaluation align with the established practices of this domain. Additionally, we propose zero-shot splits for the RVL-CDIP dataset. Furthermore, we introduce CICA (pronounced 'ki-ka'), a framework that enhances the zero-shot learning capabilities of CLIP. CICA consists of a novel 'content module' designed to leverage any generic document-related textual information. The discriminative features extracted by this module are aligned with CLIP's text and image features using a novel 'coupled-contrastive' loss. Our module improves CLIP's ZSL top-1 accuracy by 6.7% and GZSL harmonic mean by 24% on the RVL-CDIP dataset. Our module is lightweight and adds only 3.3% more parameters to CLIP. Our work sets the direction for future research in zero-shot document classification.

Tahira Shehzadi, Khurram Azeem Hashmi, Didier Stricker et al.

Table detection is the task of classifying and localizing table objects within document images. With the recent development in deep learning methods, we observe remarkable success in table detection. However, a significant amount of labeled data is required to train these models effectively. Many semi-supervised approaches are introduced to mitigate the need for a substantial amount of label data. These approaches use CNN-based detectors that rely on anchor proposals and post-processing stages such as NMS. To tackle these limitations, this paper presents a novel end-to-end semi-supervised table detection method that employs the deformable transformer for detecting table objects. We evaluate our semi-supervised method on PubLayNet, DocBank, ICADR-19 and TableBank datasets, and it achieves superior performance compared to previous methods. It outperforms the fully supervised method (Deformable transformer) by +3.4 points on 10\% labels of TableBank-both dataset and the previous CNN-based semi-supervised approach (Soft Teacher) by +1.8 points on 10\% labels of PubLayNet dataset. We hope this work opens new possibilities towards semi-supervised and unsupervised table detection methods.

Khurram Azeem Hashmi, Marcus Liwicki, Didier Stricker et al.

The first phase of table recognition is to detect the tabular area in a document. Subsequently, the tabular structures are recognized in the second phase in order to extract information from the respective cells. Table detection and structural recognition are pivotal problems in the domain of table understanding. However, table analysis is a perplexing task due to the colossal amount of diversity and asymmetry in tables. Therefore, it is an active area of research in document image analysis. Recent advances in the computing capabilities of graphical processing units have enabled deep neural networks to outperform traditional state-of-the-art machine learning methods. Table understanding has substantially benefited from the recent breakthroughs in deep neural networks. However, there has not been a consolidated description of the deep learning methods for table detection and table structure recognition. This review paper provides a thorough analysis of the modern methodologies that utilize deep neural networks. This work provided a thorough understanding of the current state-of-the-art and related challenges of table understanding in document images. Furthermore, the leading datasets and their intricacies have been elaborated along with the quantitative results. Moreover, a brief overview is given regarding the promising directions that can serve as a guide to further improve table analysis in document images.

Khurram Azeem Hashmi, Didier Stricker, Marcus Liwicki et al.

This paper presents the novel approach towards table structure recognition by leveraging the guided anchors. The concept differs from current state-of-the-art approaches for table structure recognition that naively apply object detection methods. In contrast to prior techniques, first, we estimate the viable anchors for table structure recognition. Subsequently, these anchors are exploited to locate the rows and columns in tabular images. Furthermore, the paper introduces a simple and effective method that improves the results by using tabular layouts in realistic scenarios. The proposed method is exhaustively evaluated on the two publicly available datasets of table structure recognition i.e ICDAR-2013 and TabStructDB. We accomplished state-of-the-art results on the ICDAR-2013 dataset with an average F-Measure of 95.05$\%$ (94.6$\%$ for rows and 96.32$\%$ for columns) and surpassed the baseline results on the TabStructDB dataset with an average F-Measure of 94.17$\%$ (94.08$\%$ for rows and 95.06$\%$ for columns).

Andreas Kölsch, Ashutosh Mishra, Saurabh Varshneya et al.

This paper introduces a very challenging dataset of historic German documents and evaluates Fully Convolutional Neural Network (FCNN) based methods to locate handwritten annotations of any kind in these documents. The handwritten annotations can appear in form of underlines and text by using various writing instruments, e.g., the use of pencils makes the data more challenging. We train and evaluate various end-to-end semantic segmentation approaches and report the results. The task is to classify the pixels of documents into two classes: background and handwritten annotation. The best model achieves a mean Intersection over Union (IoU) score of 95.6% on the test documents of the presented dataset. We also present a comparison of different strategies used for data augmentation and training on our presented dataset. For evaluation, we use the Layout Analysis Evaluator for the ICDAR 2017 Competition on Layout Analysis for Challenging Medieval Manuscripts.

Andreas Kölsch, Muhammad Zeshan Afzal, Markus Ebbecke et al.

This paper presents an approach for real-time training and testing for document image classification. In production environments, it is crucial to perform accurate and (time-)efficient training. Existing deep learning approaches for classifying documents do not meet these requirements, as they require much time for training and fine-tuning the deep architectures. Motivated from Computer Vision, we propose a two-stage approach. The first stage trains a deep network that works as feature extractor and in the second stage, Extreme Learning Machines (ELMs) are used for classification. The proposed approach outperforms all previously reported structural and deep learning based methods with a final accuracy of 83.24% on Tobacco-3482 dataset, leading to a relative error reduction of 25% when compared to a previous Convolutional Neural Network (CNN) based approach (DeepDocClassifier). More importantly, the training time of the ELM is only 1.176 seconds and the overall prediction time for 2,482 images is 3.066 seconds. As such, this novel approach makes deep learning-based document classification suitable for large-scale real-time applications.

Ayushman Dash, Amit Sahu, Rajveer Shringi et al.

This paper presents a novel approach, called AirScript, for creating, recognizing and visualizing documents in air. We present a novel algorithm, called 2-DifViz, that converts the hand movements in air (captured by a Myo-armband worn by a user) into a sequence of x, y coordinates on a 2D Cartesian plane, and visualizes them on a canvas. Existing sensor-based approaches either do not provide visual feedback or represent the recognized characters using prefixed templates. In contrast, AirScript stands out by giving freedom of movement to the user, as well as by providing a real-time visual feedback of the written characters, making the interaction natural. AirScript provides a recognition module to predict the content of the document created in air. To do so, we present a novel approach based on deep learning, which uses the sensor data and the visualizations created by 2-DifViz. The recognition module consists of a Convolutional Neural Network (CNN) and two Gated Recurrent Unit (GRU) Networks. The output from these three networks is fused to get the final prediction about the characters written in air. AirScript can be used in highly sophisticated environments like a smart classroom, a smart factory or a smart laboratory, where it would enable people to annotate pieces of texts wherever they want without any reference surface. We have evaluated AirScript against various well-known learning models (HMM, KNN, SVM, etc.) on the data of 12 participants. Evaluation results show that the recognition module of AirScript largely outperforms all of these models by achieving an accuracy of 91.7% in a person independent evaluation and a 96.7% accuracy in a person dependent evaluation.

Muhammad Zeshan Afzal, Andreas Kölsch, Sheraz Ahmed et al.

We present an exhaustive investigation of recent Deep Learning architectures, algorithms, and strategies for the task of document image classification to finally reduce the error by more than half. Existing approaches, such as the DeepDocClassifier, apply standard Convolutional Network architectures with transfer learning from the object recognition domain. The contribution of the paper is threefold: First, it investigates recently introduced very deep neural network architectures (GoogLeNet, VGG, ResNet) using transfer learning (from real images). Second, it proposes transfer learning from a huge set of document images, i.e. 400,000 documents. Third, it analyzes the impact of the amount of training data (document images) and other parameters to the classification abilities. We use two datasets, the Tobacco-3482 and the large-scale RVL-CDIP dataset. We achieve an accuracy of 91.13% for the Tobacco-3482 dataset while earlier approaches reach only 77.6%. Thus, a relative error reduction of more than 60% is achieved. For the large dataset RVL-CDIP, an accuracy of 90.97% is achieved, corresponding to a relative error reduction of 11.5%.

Ayushman Dash, John Cristian Borges Gamboa, Sheraz Ahmed et al.

In this work, we present the Text Conditioned Auxiliary Classifier Generative Adversarial Network, (TAC-GAN) a text to image Generative Adversarial Network (GAN) for synthesizing images from their text descriptions. Former approaches have tried to condition the generative process on the textual data; but allying it to the usage of class information, known to diversify the generated samples and improve their structural coherence, has not been explored. We trained the presented TAC-GAN model on the Oxford-102 dataset of flowers, and evaluated the discriminability of the generated images with Inception-Score, as well as their diversity using the Multi-Scale Structural Similarity Index (MS-SSIM). Our approach outperforms the state-of-the-art models, i.e., its inception score is 3.45, corresponding to a relative increase of 7.8% compared to the recently introduced StackGan. A comparison of the mean MS-SSIM scores of the training and generated samples per class shows that our approach is able to generate highly diverse images with an average MS-SSIM of 0.14 over all generated classes.

Andreas Kölsch, Muhammad Zeshan Afzal, Marcus Liwicki

In this work, we propose the combined usage of low- and high-level blocks of convolutional neural networks (CNNs) for improving object recognition. While recent research focused on either propagating the context from all layers, e.g. ResNet, (including the very low-level layers) or having multiple loss layers (e.g. GoogLeNet), the importance of the features close to the higher layers is ignored. This paper postulates that the use of context closer to the high-level layers provides the scale and translation invariance and works better than using the top layer only. In particular, we extend AlexNet and GoogLeNet by additional connections in the top $n$ layers. In order to demonstrate the effectiveness of the proposed approach, we evaluated it on the standard ImageNet task. The relative reduction of the classification error is around 1-2% without affecting the computational cost. Furthermore, we show that this approach is orthogonal to typical test data augmentation techniques, as recently introduced by Szegedy et al. (leading to a runtime reduction of 144 during test time).

Peter Burkert, Felix Trier, Muhammad Zeshan Afzal et al.

We propose a convolutional neural network (CNN) architecture for facial expression recognition. The proposed architecture is independent of any hand-crafted feature extraction and performs better than the earlier proposed convolutional neural network based approaches. We visualize the automatically extracted features which have been learned by the network in order to provide a better understanding. The standard datasets, i.e. Extended Cohn-Kanade (CKP) and MMI Facial Expression Databse are used for the quantitative evaluation. On the CKP set the current state of the art approach, using CNNs, achieves an accuracy of 99.2%. For the MMI dataset, currently the best accuracy for emotion recognition is 93.33%. The proposed architecture achieves 99.6% for CKP and 98.63% for MMI, therefore performing better than the state of the art using CNNs. Automatic facial expression recognition has a broad spectrum of applications such as human-computer interaction and safety systems. This is due to the fact that non-verbal cues are important forms of communication and play a pivotal role in interpersonal communication. The performance of the proposed architecture endorses the efficacy and reliable usage of the proposed work for real world applications.