Moongu Jeon

Daeyoung Han, Seongmin Hwang, Moongu Jeon

Conventional Generative Adversarial Networks (GANs) for Single Image Super-Resolution (SISR) often struggle with hallucinated artifacts, largely because standard discriminators evaluate overall image naturalness rather than strict conditional realism. To address this, we propose MaCo-GAN, a novel manifold-contrastive GAN framework that replaces the conventional adversarial loss with a supervised contrastive objective. A core component of our method is a dynamic fake sample synthesizer that transforms ground truth (GT) data into a spectrum of challenging, perceptually plausible fake images that strictly maintain low-resolution (LR) correspondence. Utilizing these synthesized samples, we establish a robust contrastive minimax game: the generator is trained to attract its predictions toward on-manifold fakes (low distortion) and repel them from off-manifold fakes (high distortion), while the discriminator optimizes the exact opposite. By simply replacing the adversarial loss of a baseline SR model with our proposed objective, we demonstrate consistent improvements in the perception-distortion trade-off across various benchmarks. Extensive ablation studies validate the effectiveness of our framework and provide deep insights into the dynamics of this conditional contrastive game.

Linh Van Ma, Tran Thien Dat Nguyen, Changbeom Shim et al.

This paper proposes an online visual multi-object tracking (MOT) algorithm that resolves object appearance-reappearance and occlusion. Our solution is based on the labeled random finite set (LRFS) filtering approach, which in principle, addresses disappearance, appearance, reappearance, and occlusion via a single Bayesian recursion. However, in practice, existing numerical approximations cause reappearing objects to be initialized as new tracks, especially after long periods of being undetected. In occlusion handling, the filter's efficacy is dictated by trade-offs between the sophistication of the occlusion model and computational demand. Our contribution is a novel modeling method that exploits object features to address reappearing objects whilst maintaining a linear complexity in the number of detections. Moreover, to improve the filter's occlusion handling, we propose a fuzzy detection model that takes into consideration the overlapping areas between tracks and their sizes. We also develop a fast version of the filter to further reduce the computational time. The source code is publicly available at https://github.com/linh-gist/mv-glmb-ab.

Jongmin Yu, Hyeontaek Oh, Zhongtian Sun et al.

Existing face-swapping methods often deliver competitive results in constrained settings but exhibit substantial quality degradation when handling extreme facial poses. To improve facial pose robustness, explicit geometric features are applied, but this approach remains problematic since it introduces additional dependencies and increases computational cost. Diffusion-based methods have achieved remarkable results; however, they are impractical for real-time processing. We introduce AlphaFace, which leverages an open-source vision-language model and CLIP image and text embeddings to apply novel visual and textual semantic contrastive losses. AlphaFace enables stronger identity representation and more precise attribute preservation, all while maintaining real-time performance. Comprehensive experiments across FF++, MPIE, and LPFF demonstrate that AlphaFace surpasses state-of-the-art methods in pose-challenging cases. The project is publicly available on `https://github.com/andrewyu90/Alphaface_Official.git'.

Shoaib Azam, Farzeen Munir, Ville Kyrki et al.

Learning contextual and spatial environmental representations enhances autonomous vehicle's hazard anticipation and decision-making in complex scenarios. Recent perception systems enhance spatial understanding with sensor fusion but often lack full environmental context. Humans, when driving, naturally employ neural maps that integrate various factors such as historical data, situational subtleties, and behavioral predictions of other road users to form a rich contextual understanding of their surroundings. This neural map-based comprehension is integral to making informed decisions on the road. In contrast, even with their significant advancements, autonomous systems have yet to fully harness this depth of human-like contextual understanding. Motivated by this, our work draws inspiration from human driving patterns and seeks to formalize the sensor fusion approach within an end-to-end autonomous driving framework. We introduce a framework that integrates three cameras (left, right, and center) to emulate the human field of view, coupled with top-down bird-eye-view semantic data to enhance contextual representation. The sensor data is fused and encoded using a self-attention mechanism, leading to an auto-regressive waypoint prediction module. We treat feature representation as a sequential problem, employing a vision transformer to distill the contextual interplay between sensor modalities. The efficacy of the proposed method is experimentally evaluated in both open and closed-loop settings. Our method achieves displacement error by 0.67m in open-loop settings, surpassing current methods by 6.9% on the nuScenes dataset. In closed-loop evaluations on CARLA's Town05 Long and Longest6 benchmarks, the proposed method enhances driving performance, route completion, and reduces infractions.

Labina Shrestha, Shikha Dubey, Farrukh Olimov et al.

Human action recognition is one of the challenging tasks in computer vision. The current action recognition methods use computationally expensive models for learning spatio-temporal dependencies of the action. Models utilizing RGB channels and optical flow separately, models using a two-stream fusion technique, and models consisting of both convolutional neural network (CNN) and long-short term memory (LSTM) network are few examples of such complex models. Moreover, fine-tuning such complex models is computationally expensive as well. This paper proposes a deep neural network architecture for learning such dependencies consisting of a 3D convolutional layer, fully connected (FC) layers, and attention layer, which is simpler to implement and gives a competitive performance on the UCF-101 dataset. The proposed method first learns spatial and temporal features of actions through 3D-CNN, and then the attention mechanism helps the model to locate attention to essential features for recognition.

JongHyun Park, Yechan Kim, Moongu Jeon

Recently, numerous methods have achieved impressive performance in remote sensing object detection, relying on convolution or transformer architectures. Such detectors typically have a feature backbone to extract useful features from raw input images. A common practice in current detectors is initializing the backbone with pre-trained weights available online. Fine-tuning the backbone is typically required to generate features suitable for remote-sensing images. While the prolonged training could lead to over-fitting, hindering the extraction of basic visual features, it can enable models to gradually extract deeper insights and richer representations from remote sensing data. Striking a balance between these competing factors is critical for achieving optimal performance. In this study, we aim to investigate the performance and characteristics of remote sensing object detection models under very long training schedules, and propose a novel method named Dynamic Backbone Freezing (DBF) for feature backbone fine-tuning on remote sensing object detection under long-term training. Our method addresses the dilemma of whether the backbone should extract low-level generic features or possess specific knowledge of the remote sensing domain, by introducing a module called 'Freezing Scheduler' to manage the update of backbone features during long-term training dynamically. Extensive experiments on DOTA and DIOR-R show that our approach enables more accurate model learning while substantially reducing computational costs in long-term training. Besides, it can be seamlessly adopted without additional effort due to its straightforward design. The code is available at https://github.com/unique-chan/dbf.

Farzeen Munir, Shoaib Azam, Tomasz Kucner et al.

Object detection is a core component of perception systems, providing the ego vehicle with information about its surroundings to ensure safe route planning. While cameras and Lidar have significantly advanced perception systems, their performance can be limited in adverse weather conditions. In contrast, millimeter-wave technology enables radars to function effectively in such conditions. However, relying solely on radar for building a perception system doesn't fully capture the environment due to the data's sparse nature. To address this, sensor fusion strategies have been introduced. We propose a dual-branch framework to integrate radar and Lidar data for enhanced object detection. The primary branch focuses on extracting radar features, while the auxiliary branch extracts Lidar features. These are then combined using additive attention. Subsequently, the integrated features are processed through a novel Parallel Forked Structure (PFS) to manage scale variations. A region proposal head is then utilized for object detection. We evaluated the effectiveness of our proposed method on the Radiate dataset using COCO metrics. The results show that it surpasses state-of-the-art methods by $1.89\%$ and $2.61\%$ in favorable and adverse weather conditions, respectively. This underscores the value of radar-Lidar fusion in achieving precise object detection and localization, especially in challenging weather conditions.

Seongmin Hwang, Daeyoung Han, Cheolkon Jung et al.

The surge in interest regarding image dehazing has led to notable advancements in deep learning-based single image dehazing approaches, exhibiting impressive performance in recent studies. Despite these strides, many existing methods fall short in meeting the efficiency demands of practical applications. In this paper, we introduce WaveDH, a novel and compact ConvNet designed to address this efficiency gap in image dehazing. Our WaveDH leverages wavelet sub-bands for guided up-and-downsampling and frequency-aware feature refinement. The key idea lies in utilizing wavelet decomposition to extract low-and-high frequency components from feature levels, allowing for faster processing while upholding high-quality reconstruction. The downsampling block employs a novel squeeze-and-attention scheme to optimize the feature downsampling process in a structurally compact manner through wavelet domain learning, preserving discriminative features while discarding noise components. In our upsampling block, we introduce a dual-upsample and fusion mechanism to enhance high-frequency component awareness, aiding in the reconstruction of high-frequency details. Departing from conventional dehazing methods that treat low-and-high frequency components equally, our feature refinement block strategically processes features with a frequency-aware approach. By employing a coarse-to-fine methodology, it not only refines the details at frequency levels but also significantly optimizes computational costs. The refinement is performed in a maximum 8x downsampled feature space, striking a favorable efficiency-vs-accuracy trade-off. Extensive experiments demonstrate that our method, WaveDH, outperforms many state-of-the-art methods on several image dehazing benchmarks with significantly reduced computational costs. Our code is available at https://github.com/AwesomeHwang/WaveDH.

Yechan Kim, SooYeon Kim, Moongu Jeon

Data augmentation has shown significant advancements in computer vision to improve model performance over the years, particularly in scenarios with limited and insufficient data. Currently, most studies focus on adjusting the image or its features to expand the size, quality, and variety of samples during training in various tasks including object detection. However, we argue that it is necessary to investigate bounding box transformations as a data augmentation technique rather than image-level transformations, especially in aerial imagery due to potentially inconsistent bounding box annotations. Hence, this letter presents a thorough investigation of bounding box transformation in terms of scaling, rotation, and translation for remote sensing object detection. We call this augmentation strategy NBBOX (Noise Injection into Bounding Box). We conduct extensive experiments on DOTA and DIOR-R, both well-known datasets that include a variety of rotated generic objects in aerial images. Experimental results show that our approach significantly improves remote sensing object detection without whistles and bells and it is more time-efficient than other state-of-the-art augmentation strategies.

Zubia Naz, Farhan Asghar, Muhammad Ishfaq Hussain et al.

Automated medical image captioning translates complex radiological images into diagnostic narratives that can support reporting workflows. We present a Swin-BART encoder-decoder system with a lightweight regional attention module that amplifies diagnostically salient regions before cross-attention. Trained and evaluated on ROCO, our model achieves state-of-the-art semantic fidelity while remaining compact and interpretable. We report results as mean$\pm$std over three seeds and include $95\%$ confidence intervals. Compared with baselines, our approach improves ROUGE (proposed 0.603, ResNet-CNN 0.356, BLIP2-OPT 0.255) and BERTScore (proposed 0.807, BLIP2-OPT 0.645, ResNet-CNN 0.623), with competitive BLEU, CIDEr, and METEOR. We further provide ablations (regional attention on/off and token-count sweep), per-modality analysis (CT/MRI/X-ray), paired significance tests, and qualitative heatmaps that visualize the regions driving each description. Decoding uses beam search (beam size $=4$), length penalty $=1.1$, $no\_repeat\_ngram\_size$ $=3$, and max length $=128$. The proposed design yields accurate, clinically phrased captions and transparent regional attributions, supporting safe research use with a human in the loop.

Seongmin Hwang, Daeyoung Han, Moongu Jeon

Multispectral object detection aims to leverage complementary information from visible (RGB) and infrared (IR) modalities to enable robust performance under diverse environmental conditions. Our key insight, derived from wavelet analysis and empirical observations, is that IR images contain structurally rich high-frequency information critical for object detection, making an infrared-centric approach highly effective. To capitalize on this finding, we propose Infrared-Centric Fusion (IC-Fusion), a lightweight and modality-aware sensor fusion method that prioritizes infrared features while effectively integrating complementary RGB semantic context. IC-Fusion adopts a compact RGB backbone and designs a novel fusion module comprising a Multi-Scale Feature Distillation (MSFD) block to enhance RGB features and a three-stage fusion block with a Cross-Modal Channel Shuffle Gate (CCSG), a Cross-Modal Large Kernel Gate (CLKG), and a Channel Shuffle Projection (CSP) to facilitate effective cross-modal interaction. Experiments on the FLIR and LLVIP benchmarks demonstrate the superior effectiveness and efficiency of our IR-centric fusion strategy, further validating its benefits. Our code is available at https://github.com/smin-hwang/IC-Fusion.

Linh Van Ma, Muhammad Ishfaq Hussain, Kin-Choong Yow et al.

The MS-GLMB filter offers a robust framework for tracking multiple objects through the use of multi-sensor data. Building on this, the MV-GLMB and MV-GLMB-AB filters enhance the MS-GLMB capabilities by employing cameras for 3D multi-sensor multi-object tracking, effectively addressing occlusions. However, both filters depend on overlapping fields of view from the cameras to combine complementary information. In this paper, we introduce an improved approach that integrates an additional sensor, such as LiDAR, into the MS-GLMB framework for 3D multi-object tracking. Specifically, we present a new LiDAR measurement model, along with a multi-camera and LiDAR multi-object measurement model. Our experimental results demonstrate a significant improvement in tracking performance compared to existing MS-GLMB-based methods. Importantly, our method eliminates the need for overlapping fields of view, broadening the applicability of the MS-GLMB filter. Our source code for nuScenes dataset is available at https://github.com/linh-gist/ms-glmb-nuScenes.

Seongmin Hwang, Daeyoung Han, Moongu Jeon

End-to-end Transformer-based detectors (DETRs) have demonstrated strong detection performance. However, domain generalization (DG) research has primarily focused on convolutional neural network (CNN)-based detectors, while paying little attention to enhancing the robustness of DETRs. In this letter, we introduce a Domain Generalized DEtection TRansformer (DG-DETR), a simple, effective, and plug-and-play method that improves out-of-distribution (OOD) robustness for DETRs. Specifically, we propose a novel domain-agnostic query selection strategy that removes domain-induced biases from object queries via orthogonal projection onto the instance-specific style space. Additionally, we leverage a wavelet decomposition to disentangle features into domain-invariant and domain-specific components, enabling synthesis of diverse latent styles while preserving the semantic features of objects. Experimental results validate the effectiveness of DG-DETR. Our code is available at https://github.com/sminhwang/DG-DETR.

Linh Van Ma, Tin Trung Tran, Moongu Jeon

Most Gaze estimation research only works on a setup condition that a camera perfectly captures eyes gaze. They have not literarily specified how to set up a camera correctly for a given position of a person. In this paper, we carry out a study on gaze estimation with a logical camera setup position. We further bring our research in a practical application by using inexpensive edge devices with a realistic scenario. That is, we first set up a shopping environment where we want to grasp customers gazing behaviors. This setup needs an optimal camera position in order to maintain estimation accuracy from existing gaze estimation research. We then apply the state-of-the-art of few-shot learning gaze estimation to reduce training sampling in the inference phase. In the experiment, we perform our implemented research on NVIDIA Jetson TX2 and achieve a reasonable speed, 12 FPS which is faster compared with our reference work, without much degradation of gaze estimation accuracy. The source code is released at https://github.com/linh-gist/GazeEstimationTX2.

Yeongmin Ko, Younkwan Lee, Shoaib Azam et al.

Perception techniques for autonomous driving should be adaptive to various environments. In the case of traffic line detection, an essential perception module, many condition should be considered, such as number of traffic lines and computing power of the target system. To address these problems, in this paper, we propose a traffic line detection method called Point Instance Network (PINet); the method is based on the key points estimation and instance segmentation approach. The PINet includes several stacked hourglass networks that are trained simultaneously. Therefore the size of the trained models can be chosen according to the computing power of the target environment. We cast a clustering problem of the predicted key points as an instance segmentation problem; the PINet can be trained regardless of the number of the traffic lines. The PINet achieves competitive accuracy and false positive on the TuSimple and Culane datasets, popular public datasets for lane detection. Our code is available at https://github.com/koyeongmin/PINet_new

Young-Chul Yoon, Du Yong Kim, Young-min Song et al.

In online multi-target tracking, modeling of appearance and geometric similarities between pedestrians visual scenes is of great importance. The higher dimension of inherent information in the appearance model compared to the geometric model is problematic in many ways. However, due to the recent success of deep-learning-based methods, handling of high-dimensional appearance information becomes feasible. Among many deep neural networks, Siamese network with triplet loss has been widely adopted as an effective appearance feature extractor. Since the Siamese network can extract the features of each input independently, one can update and maintain target-specific features. However, it is not suitable for multi-target settings that require comparison with other inputs. To address this issue, we propose a novel track appearance model based on the joint-inference network. The proposed method enables a comparison of two inputs to be used for adaptive appearance modeling and contributes to the disambiguation of target-observation matching and to the consolidation of identity consistency. Diverse experimental results support the effectiveness of our method. Our work was recognized as the 3rd-best tracker in BMTT MOTChallenge 2019, held at CVPR2019. The code is available at https://github.com/yyc9268/Deep-TAMA.

Linh Van Ma, Tran Thien Dat Nguyen, Moongu Jeon

This paper proposes a fast and online method for jointly performing 3D multi-object tracking and pose estimation using multiple monocular cameras. Our algorithm requires only 2D bounding box and pose detections, eliminating the need for costly 3D training data or computationally expensive deep learning models. Our solution is an efficient implementation of a Bayes-optimal multi-object tracking filter, enhancing computational efficiency while maintaining accuracy. We demonstrate that our algorithm is significantly faster than state-of-the-art methods without compromising accuracy, using only publicly available pre-trained 2D detection models. We also illustrate the robust performance of our algorithm in scenarios where multiple cameras are intermittently disconnected or reconnected during operation.

Linh Van Ma, Muhammad Ishfaq Hussain, JongHyun Park et al.

We investigate the application of ByteTrack in the realm of multiple object tracking. ByteTrack, a simple tracking algorithm, enables the simultaneous tracking of multiple objects by strategically incorporating detections with a low confidence threshold. Conventionally, objects are initially associated with high confidence threshold detections. When the association between objects and detections becomes ambiguous, ByteTrack extends the association to lower confidence threshold detections. One notable drawback of the existing ByteTrack approach is its reliance on a fixed threshold to differentiate between high and low-confidence detections. In response to this limitation, we introduce a novel and adaptive approach. Our proposed method entails a dynamic adjustment of the confidence threshold, leveraging insights derived from overall detections. Through experimentation, we demonstrate the effectiveness of our adaptive confidence threshold technique while maintaining running time compared to ByteTrack.

Yechan Kim, Dongho Yoon, Younkwan Lee et al.

While previous studies on image segmentation focus on handling severe (or explicit) label noise, real-world datasets also exhibit subtle (or implicit) label imperfections. These arise from inherent challenges, such as ambiguous object boundaries and annotator variability. Although not explicitly present, such mild and latent noise can still impair model performance. Typical data augmentation methods, which apply identical transformations to the image and its label, risk amplifying these subtle imperfections and limiting the model's generalization capacity. In this paper, we introduce NSegment+, a novel augmentation framework that decouples image and label transformations to address such realistic noise for semantic segmentation. By introducing controlled elastic deformations only to segmentation labels while preserving the original images, our method encourages models to focus on learning robust representations of object structures despite minor label inconsistencies. Extensive experiments demonstrate that NSegment+ consistently improves performance, achieving mIoU gains of up to +2.29, +2.38, +1.75, and +3.39 in average on Vaihingen, LoveDA, Cityscapes, and PASCAL VOC, respectively-even without bells and whistles, highlighting the importance of addressing implicit label noise. These gains can be further amplified when combined with other training tricks, including CutMix and Label Smoothing.

Muhammad Ishfaq Hussain, Ma Van Linh, Zubia Naz et al.

Enhancing scene understanding in adverse visibility conditions remains a critical challenge for surveillance and autonomous navigation systems. Conventional imaging modalities, such as RGB and thermal infrared (MWIR / LWIR), when fused, often struggle to deliver comprehensive scene information, particularly under conditions of atmospheric interference or inadequate illumination. To address these limitations, Short-Wave Infrared (SWIR) imaging has emerged as a promising modality due to its ability to penetrate atmospheric disturbances and differentiate materials with improved clarity. However, the advancement and widespread implementation of SWIR-based systems face significant hurdles, primarily due to the scarcity of publicly accessible SWIR datasets. In response to this challenge, our research introduces an approach to synthetically generate SWIR-like structural/contrast cues (without claiming spectral reproduction) images from existing LWIR data using advanced contrast enhancement techniques. We then propose a multimodal fusion framework integrating synthetic SWIR, LWIR, and RGB modalities, employing an optimized encoder-decoder neural network architecture with modality-specific encoders and a softmax-gated fusion head. Comprehensive experiments on public RGB-LWIR benchmarks (M3FD, TNO, CAMEL, MSRS, RoadScene) and an additional private real RGB-MWIR-SWIR dataset demonstrate that our synthetic-SWIR-enhanced fusion framework improves fused-image quality (contrast, edge definition, structural fidelity) while maintaining real-time performance. We also add fair trimodal baselines (LP, LatLRR, GFF) and cascaded trimodal variants of U2Fusion/SwinFusion under a unified protocol. The outcomes highlight substantial potential for real-world applications in surveillance and autonomous systems.

Yechan Kim, DongHo Yoon, SooYeon Kim et al.

Labeling errors in remote sensing (RS) image segmentation datasets often remain implicit and subtle due to ambiguous class boundaries, mixed pixels, shadows, complex terrain features, and subjective annotator bias. Furthermore, the scarcity of annotated RS data due to the high cost of labeling complicates training noise-robust models. While sophisticated mechanisms such as label selection or noise correction might address the issue mentioned above, they tend to increase training time and add implementation complexity. In this paper, we propose NSegment-a simple yet effective data augmentation solution to mitigate this issue. Unlike traditional methods, it applies elastic transformations only to segmentation labels, varying deformation intensity per sample in each training epoch to address annotation inconsistencies. Experimental results demonstrate that our approach improves the performance of RS image segmentation over various state-of-the-art models.

Hyeonsoo Jang, Yeongmin Ko, Younkwan Lee et al.

Depth estimation plays a important role in SLAM, odometry, and autonomous driving. Especially, monocular depth estimation is profitable technology because of its low cost, memory, and computation. However, it is not a sufficiently predicting depth map due to a camera often failing to get a clean image because of light conditions. To solve this problem, various sensor fusion method has been proposed. Even though it is a powerful method, sensor fusion requires expensive sensors, additional memory, and high computational performance. In this paper, we present color image and monochrome image pixel-level fusion and stereo matching with partially enhanced correlation coefficient maximization. Our methods not only outperform the state-of-the-art works across all metrics but also efficient in terms of cost, memory, and computation. We also validate the effectiveness of our design with an ablation study.

Farzeen Munir, Shoaib Azam, Byung-Geun Lee et al.

Imitation learning is employed to learn sensorimotor coordination for steering angle prediction in an end-to-end fashion requires expert demonstrations. These expert demonstrations are paired with environmental perception and vehicle control data. The conventional frame-based RGB camera is the most common exteroceptive sensor modality used to acquire the environmental perception data. The frame-based RGB camera has produced promising results when used as a single modality in learning end-to-end lateral control. However, the conventional frame-based RGB camera has limited operability in illumination variation conditions and is affected by the motion blur. The event camera provides complementary information to the frame-based RGB camera. This work explores the fusion of frame-based RGB and event data for learning end-to-end lateral control by predicting steering angle. In addition, how the representation from event data fuse with frame-based RGB data helps to predict the lateral control robustly for the autonomous vehicle. To this end, we propose DRFuser, a novel convolutional encoder-decoder architecture for learning end-to-end lateral control. The encoder module is branched between the frame-based RGB data and event data along with the self-attention layers. Moreover, this study has also contributed to our own collected dataset comprised of event, frame-based RGB, and vehicle control data. The efficacy of the proposed method is experimentally evaluated on our collected dataset, Davis Driving dataset (DDD), and Carla Eventscape dataset. The experimental results illustrate that the proposed method DRFuser outperforms the state-of-the-art in terms of root-mean-square error (RMSE) and mean absolute error (MAE) used as evaluation metrics.

Shraman Pal, Mansi Uniyal, Jihong Park et al.

In recent years, there have been great advances in the field of decentralized learning with private data. Federated learning (FL) and split learning (SL) are two spearheads possessing their pros and cons, and are suited for many user clients and large models, respectively. To enjoy both benefits, hybrid approaches such as SplitFed have emerged of late, yet their fundamentals have still been illusive. In this work, we first identify the fundamental bottlenecks of SL, and thereby propose a scalable SL framework, coined SGLR. The server under SGLR broadcasts a common gradient averaged at the split-layer, emulating FL without any additional communication across clients as opposed to SplitFed. Meanwhile, SGLR splits the learning rate into its server-side and client-side rates, and separately adjusts them to support many clients in parallel. Simulation results corroborate that SGLR achieves higher accuracy than other baseline SL methods including SplitFed, which is even on par with FL consuming higher energy and communication costs. As a secondary result, we observe greater reduction in leakage of sensitive information via mutual information using SLGR over the baselines.

Farzeen Munir, Shoaib Azam, Unse Fatima et al.

The research advancements have made the neural network algorithms deployed in the autonomous vehicle to perceive the surrounding. The standard exteroceptive sensors that are utilized for the perception of the environment are cameras and Lidar. Therefore, the neural network algorithms developed using these exteroceptive sensors have provided the necessary solution for the autonomous vehicle's perception. One major drawback of these exteroceptive sensors is their operability in adverse weather conditions, for instance, low illumination and night conditions. The useability and affordability of thermal cameras in the sensor suite of the autonomous vehicle provide the necessary improvement in the autonomous vehicle's perception in adverse weather conditions. The semantics of the environment benefits the robust perception, which can be achieved by segmenting different objects in the scene. In this work, we have employed the thermal camera for semantic segmentation. We have designed an attention-based Recurrent Convolution Network (RCNN) encoder-decoder architecture named ARTSeg for thermal semantic segmentation. The main contribution of this work is the design of encoder-decoder architecture, which employ units of RCNN for each encoder and decoder block. Furthermore, additive attention is employed in the decoder module to retain high-resolution features and improve the localization of features. The efficacy of the proposed method is evaluated on the available public dataset, showing better performance with other state-of-the-art methods in mean intersection over union (IoU).

Younkwan Lee, Jihyo Jeon, Yeongmin Ko et al.

Visual perception in autonomous driving is a crucial part of a vehicle to navigate safely and sustainably in different traffic conditions. However, in bad weather such as heavy rain and haze, the performance of visual perception is greatly affected by several degrading effects. Recently, deep learning-based perception methods have addressed multiple degrading effects to reflect real-world bad weather cases but have shown limited success due to 1) high computational costs for deployment on mobile devices and 2) poor relevance between image enhancement and visual perception in terms of the model ability. To solve these issues, we propose a task-driven image enhancement network connected to the high-level vision task, which takes in an image corrupted by bad weather as input. Specifically, we introduce a novel low memory network to reduce most of the layer connections of dense blocks for less memory and computational cost while maintaining high performance. We also introduce a new task-driven training strategy to robustly guide the high-level task model suitable for both high-quality restoration of images and highly accurate perception. Experiment results demonstrate that the proposed method improves the performance among lane and 2D object detection, and depth estimation largely under adverse weather in terms of both low memory and accuracy.

Shikha Dubey, Farrukh Olimov, Muhammad Aasim Rafique et al.

Automatic transcription of scene understanding in images and videos is a step towards artificial general intelligence. Image captioning is a nomenclature for describing meaningful information in an image using computer vision techniques. Automated image captioning techniques utilize encoder and decoder architecture, where the encoder extracts features from an image and the decoder generates a transcript. In this work, we investigate two unexplored ideas for image captioning using transformers: First, we demonstrate the enforcement of using objects' relevance in the surrounding environment. Second, learning an explicit association between labels and language constructs. We propose label-attention Transformer with geometrically coherent objects (LATGeO). The proposed technique acquires a proposal of geometrically coherent objects using a deep neural network (DNN) and generates captions by investigating their relationships using a label-attention module. Object coherence is defined using the localized ratio of the geometrical properties of the proposals. The label-attention module associates the extracted objects classes to the available dictionary using self-attention layers. The experimentation results show that objects' relevance in surroundings and binding of their visual feature with their geometrically localized ratios combined with its associated labels help in defining meaningful captions. The proposed framework is tested on the MSCOCO dataset, and a thorough evaluation resulting in overall better quantitative scores pronounces its superiority.

Muhamamd Ishfaq Hussain, Zubia Naz, Muhammad Aasim Rafique et al.

Depth perception is crucial for spatial understanding and has traditionally been achieved through stereoscopic imaging. However, the precision of depth estimation using stereoscopic methods depends on the accurate calibration of binocular vision sensors. Monocular cameras, while more accessible, often suffer from reduced accuracy, especially under challenging imaging conditions. Optical sensors, too, face limitations in adverse environments, leading researchers to explore radar technology as a reliable alternative. Although radar provides coarse but accurate signals, its integration with fine-grained monocular camera data remains underexplored. In this research, we propose DepthSense, a novel radar-assisted monocular depth enhancement approach. DepthSense employs an encoder-decoder architecture, a Radar Residual Network, feature fusion with a spatial attention mechanism, and an ordinal regression layer to deliver precise depth estimations. We conducted extensive experiments on the nuScenes dataset to validate the effectiveness of DepthSense. Our methodology not only surpasses existing approaches in quantitative performance but also reduces parameter complexity and inference times. Our findings demonstrate that DepthSense represents a significant advancement over traditional stereo methods, offering a robust and efficient solution for depth estimation in autonomous driving. By leveraging the complementary strengths of radar and monocular camera data, DepthSense sets a new benchmark in the field, paving the way for more reliable and accurate spatial perception systems.

Farzeen Munir, Shoaib Azam, Moongu Jeon

The sensibility and sensitivity of the environment play a decisive role in the safe and secure operation of autonomous vehicles. This perception of the surrounding is way similar to human visual representation. The human's brain perceives the environment by utilizing different sensory channels and develop a view-invariant representation model. Keeping in this context, different exteroceptive sensors are deployed on the autonomous vehicle for perceiving the environment. The most common exteroceptive sensors are camera, Lidar and radar for autonomous vehicle's perception. Despite being these sensors have illustrated their benefit in the visible spectrum domain yet in the adverse weather conditions, for instance, at night, they have limited operation capability, which may lead to fatal accidents. In this work, we explore thermal object detection to model a view-invariant model representation by employing the self-supervised contrastive learning approach. For this purpose, we have proposed a deep neural network Self Supervised Thermal Network (SSTN) for learning the feature embedding to maximize the information between visible and infrared spectrum domain by contrastive learning, and later employing these learned feature representation for the thermal object detection using multi-scale encoder-decoder transformer network. The proposed method is extensively evaluated on the two publicly available datasets: the FLIR-ADAS dataset and the KAIST Multi-Spectral dataset. The experimental results illustrate the efficacy of the proposed method.

Farrukh Olimov, Shikha Dubey, Labina Shrestha et al.

Real-time image captioning, along with adequate precision, is the main challenge of this research field. The present work, Multiple Transformers for Self-Attention Mechanism (MTSM), utilizes multiple transformers to address these problems. The proposed algorithm, MTSM, acquires region proposals using a transformer detector (DETR). Consequently, MTSM achieves the self-attention mechanism by transferring these region proposals and their visual and geometrical features through another transformer and learns the objects' local and global interconnections. The qualitative and quantitative results of the proposed algorithm, MTSM, are shown on the MSCOCO dataset.

Shoaib Azam, Farzeen Munir, Moongu Jeon

Autonomous vehicles are conceived to provide safe and secure services by validating the safety standards as indicated by SOTIF-ISO/PAS-21448 (Safety of the intended functionality). Keeping in this context, the perception of the environment plays an instrumental role in conjunction with localization, planning and control modules. As a pivotal algorithm in the perception stack, object detection provides extensive insights into the autonomous vehicle's surroundings. Camera and Lidar are extensively utilized for object detection among different sensor modalities, but these exteroceptive sensors have limitations in resolution and adverse weather conditions. In this work, radar-based object detection is explored provides a counterpart sensor modality to be deployed and used in adverse weather conditions. The radar gives complex data; for this purpose, a channel boosting feature ensemble method with transformer encoder-decoder network is proposed. The object detection task using radar is formulated as a set prediction problem and evaluated on the publicly available dataset in both good and good-bad weather conditions. The proposed method's efficacy is extensively evaluated using the COCO evaluation metric, and the best-proposed model surpasses its state-of-the-art counterpart method by $12.55\%$ and $12.48\%$ in both good and good-bad weather conditions.

Farzeen Munir, Shoaib Azam, Moongu Jeon et al.

Modern vehicles are equipped with various driver-assistance systems, including automatic lane keeping, which prevents unintended lane departures. Traditional lane detection methods incorporate handcrafted or deep learning-based features followed by postprocessing techniques for lane extraction using frame-based RGB cameras. The utilization of frame-based RGB cameras for lane detection tasks is prone to illumination variations, sun glare, and motion blur, which limits the performance of lane detection methods. Incorporating an event camera for lane detection tasks in the perception stack of autonomous driving is one of the most promising solutions for mitigating challenges encountered by frame-based RGB cameras. The main contribution of this work is the design of the lane marking detection model, which employs the dynamic vision sensor. This paper explores the novel application of lane marking detection using an event camera by designing a convolutional encoder followed by the attention-guided decoder. The spatial resolution of the encoded features is retained by a dense atrous spatial pyramid pooling (ASPP) block. The additive attention mechanism in the decoder improves performance for high dimensional input encoded features that promote lane localization and relieve postprocessing computation. The efficacy of the proposed work is evaluated using the DVS dataset for lane extraction (DET). The experimental results show a significant improvement of $5.54\%$ and $5.03\%$ in $F1$ scores in multiclass and binary-class lane marking detection tasks. Additionally, the intersection over union ($IoU$) scores of the proposed method surpass those of the best-performing state-of-the-art method by $6.50\%$ and $9.37\%$ in multiclass and binary-class tasks, respectively.

Yechan Kim, Younkwan Lee, Moongu Jeon

Recently, deep learning models have achieved great success in computer vision applications, relying on large-scale class-balanced datasets. However, imbalanced class distributions still limit the wide applicability of these models due to degradation in performance. To solve this problem, in this paper, we concentrate on the study of cross entropy which mostly ignores output scores on incorrect classes. This work discovers that neutralizing predicted probabilities on incorrect classes improves the prediction accuracy for imbalanced image classification. This paper proposes a simple but effective loss named complement cross entropy based on this finding. The proposed loss makes the ground truth class overwhelm the other classes in terms of softmax probability, by neutralizing probabilities of incorrect classes, without additional training procedures. Along with it, this loss facilitates the models to learn key information especially from samples on minority classes. It ensures more accurate and robust classification results on imbalanced distributions. Extensive experiments on imbalanced datasets demonstrate the effectiveness of the proposed method.

Young-min Song, Young-chul Yoon, Kwangjin Yoon et al.

In this paper, we propose a highly practical fully online multi-object tracking and segmentation (MOTS) method that uses instance segmentation results as an input. The proposed method is based on the Gaussian mixture probability hypothesis density (GMPHD) filter, a hierarchical data association (HDA), and a mask-based affinity fusion (MAF) model to achieve high-performance online tracking. The HDA consists of two associations: segment-to-track and track-to-track associations. One affinity, for position and motion, is computed by using the GMPHD filter, and the other affinity, for appearance is computed by using the responses from a single object tracker such as a kernalized correlation filter. These two affinities are simply fused by using a score-level fusion method such as min-max normalization referred to as MAF. In addition, to reduce the number of false positive segments, we adopt mask IoU-based merging (mask merging). The proposed MOTS framework with the key modules: HDA, MAF, and mask merging, is easily extensible to simultaneously track multiple types of objects with CPU only execution in parallel processing. In addition, the developed framework only requires simple parameter tuning unlike many existing MOTS methods that need intensive hyperparameter optimization. In the experiments on the two popular MOTS datasets, the key modules show some improvements. For instance, ID-switch decreases by more than half compared to a baseline method in the training sets. In conclusion, our tracker achieves state-of-the-art MOTS performance in the test sets.

Farzeen Munir, Shoaib Azam, Muhammd Aasim Rafique et al.

Underexposure regions are vital to construct a complete perception of the surroundings for safe autonomous driving. The availability of thermal cameras has provided an essential alternate to explore regions where other optical sensors lack in capturing interpretable signals. A thermal camera captures an image using the heat difference emitted by objects in the infrared spectrum, and object detection in thermal images becomes effective for autonomous driving in challenging conditions. Although object detection in the visible spectrum domain imaging has matured, thermal object detection lacks effectiveness. A significant challenge is scarcity of labeled data for the thermal domain which is desiderata for SOTA artificial intelligence techniques. This work proposes a domain adaptation framework which employs a style transfer technique for transfer learning from visible spectrum images to thermal images. The framework uses a generative adversarial network (GAN) to transfer the low-level features from the visible spectrum domain to the thermal domain through style consistency. The efficacy of the proposed method of object detection in thermal images is evident from the improved results when used styled images from publicly available thermal image datasets (FLIR ADAS and KAIST Multi-Spectral).

Younkwan Lee, Jihyo Jeon, Jongmin Yu et al.

Traffic scene recognition, which requires various visual classification tasks, is a critical ingredient in autonomous vehicles. However, most existing approaches treat each relevant task independently from one another, never considering the entire system as a whole. Because of this, they are limited to utilizing a task-specific set of features for all possible tasks of inference-time, which ignores the capability to leverage common task-invariant contextual knowledge for the task at hand. To address this problem, we propose an algorithm to jointly learn the task-specific and shared representations by adopting a multi-task learning network. Specifically, we present a lower bound for the mutual information constraint between shared feature embedding and input that is considered to be able to extract common contextual information across tasks while preserving essential information of each task jointly. The learned representations capture richer contextual information without additional task-specific network. Extensive experiments on the large-scale dataset HSD demonstrate the effectiveness and superiority of our network over state-of-the-art methods.

Jongmin Yu, Yongsang Yoon, Moongu Jeon

In skeleton-based action recognition, graph convolutional networks (GCNs), which model human body skeletons using graphical components such as nodes and connections, have achieved remarkable performance recently. However, current state-of-the-art methods for skeleton-based action recognition usually work on the assumption that the completely observed skeletons will be provided. This may be problematic to apply this assumption in real scenarios since there is always a possibility that captured skeletons are incomplete or noisy. In this work, we propose a skeleton-based action recognition method which is robust to noise information of given skeleton features. The key insight of our approach is to train a model by maximizing the mutual information between normal and noisy skeletons using a predictive coding manner. We have conducted comprehensive experiments about skeleton-based action recognition with defected skeletons using NTU-RGB+D and Kinetics-Skeleton datasets. The experimental results demonstrate that our approach achieves outstanding performance when skeleton samples are noised compared with existing state-of-the-art methods.

Shikha Dubey, Abhijeet Boragule, Moongu Jeon

Abnormal activity detection is one of the most challenging tasks in the field of computer vision. This study is motivated by the recent state-of-art work of abnormal activity detection, which utilizes both abnormal and normal videos in learning abnormalities with the help of multiple instance learning by providing the data with video-level information. In the absence of temporal-annotations, such a model is prone to give a false alarm while detecting the abnormalities. For this reason, in this paper, we focus on the task of minimizing the false alarm rate while performing an abnormal activity detection task. The mitigation of these false alarms and recent advancement of 3D deep neural network in video action recognition task collectively give us motivation to exploit the 3D ResNet in our proposed method, which helps to extract spatial-temporal features from the videos. Afterwards, using these features and deep multiple instance learning along with the proposed ranking loss, our model learns to predict the abnormality score at the video segment level. Therefore, our proposed method 3D deep Multiple Instance Learning with ResNet (MILR) along with the new proposed ranking loss function achieves the best performance on the UCF-Crime benchmark dataset, as compared to other state-of-art methods. The effectiveness of our proposed method is demonstrated on the UCF-Crime dataset.

Jongmin Yu, Duyong Kim, Younkwan Lee et al.

In the past few years, the performance of road defect detection has been remarkably improved thanks to advancements on various studies on computer vision and deep learning. Although a large-scale and well-annotated datasets enhance the performance of detecting road pavement defects to some extent, it is still challengeable to derive a model which can perform reliably for various road conditions in practice, because it is intractable to construct a dataset considering diverse road conditions and defect patterns. To end this, we propose an unsupervised approach to detecting road defects, using Adversarial Image-to-Frequency Transform (AIFT). AIFT adopts the unsupervised manner and adversarial learning in deriving the defect detection model, so AIFT does not need annotations for road pavement defects. We evaluate the efficiency of AIFT using GAPs384 dataset, Cracktree200 dataset, CRACK500 dataset, and CFD dataset. The experimental results demonstrate that the proposed approach detects various road detects, and it outperforms existing state-of-the-art approaches.

Jongmin Yu, Sangwoo Park, Sangwook Lee et al.

We propose a condition-adaptive representation learning framework for the driver drowsiness detection based on 3D-deep convolutional neural network. The proposed framework consists of four models: spatio-temporal representation learning, scene condition understanding, feature fusion, and drowsiness detection. The spatio-temporal representation learning extracts features that can describe motions and appearances in video simultaneously. The scene condition understanding classifies the scene conditions related to various conditions about the drivers and driving situations such as statuses of wearing glasses, illumination condition of driving, and motion of facial elements such as head, eye, and mouth. The feature fusion generates a condition-adaptive representation using two features extracted from above models. The detection model recognizes drivers drowsiness status using the condition-adaptive representation. The condition-adaptive representation learning framework can extract more discriminative features focusing on each scene condition than the general representation so that the drowsiness detection method can provide more accurate results for the various driving situations. The proposed framework is evaluated with the NTHU Drowsy Driver Detection video dataset. The experimental results show that our framework outperforms the existing drowsiness detection methods based on visual analysis.

Younkwan Lee, Juhyun Lee, Yoojin Hong et al.

Recent road marking recognition has achieved great success in the past few years along with the rapid development of deep learning. Although considerable advances have been made, they are often over-dependent on unrepresentative datasets and constrained conditions. In this paper, to overcome these drawbacks, we propose an alternative method that achieves higher accuracy and generates high-quality samples as data augmentation. With the following two major contributions: 1) The proposed deblurring network can successfully recover a clean road marking from a blurred one by adopting generative adversarial networks (GAN). 2) The proposed data augmentation method, based on mutual information, can preserve and learn semantic context from the given dataset. We construct and train a class-conditional GAN to increase the size of training set, which makes it suitable to recognize target. The experimental results have shown that our proposed framework generates deblurred clean samples from blurry ones, and outperforms other methods even with unconstrained road marking datasets.

Younkwan Lee, Jiwon Jun, Yoojin Hong et al.

Although most current license plate (LP) recognition applications have been significantly advanced, they are still limited to ideal environments where training data are carefully annotated with constrained scenes. In this paper, we propose a novel license plate recognition method to handle unconstrained real world traffic scenes. To overcome these difficulties, we use adversarial super-resolution (SR), and one-stage character segmentation and recognition. Combined with a deep convolutional network based on VGG-net, our method provides simple but reasonable training procedure. Moreover, we introduce GIST-LP, a challenging LP dataset where image samples are effectively collected from unconstrained surveillance scenes. Experimental results on AOLP and GIST-LP dataset illustrate that our method, without any scene-specific adaptation, outperforms current LP recognition approaches in accuracy and provides visual enhancement in our SR results that are easier to understand than original data.

Younkwan Lee, Juhyun Lee, Hoyeon Ahn et al.

In this paper, we present an algorithm for real-world license plate recognition (LPR) from a low-quality image. Our method is built upon a framework that includes denoising and rectification, and each task is conducted by Convolutional Neural Networks. Existing denoising and rectification have been treated separately as a single network in previous research. In contrast to the previous work, we here propose an end-to-end trainable network for image recovery, Single Noisy Image DEnoising and Rectification (SNIDER), which focuses on solving both the problems jointly. It overcomes those obstacles by designing a novel network to address the denoising and rectification jointly. Moreover, we propose a way to leverage optimization with the auxiliary tasks for multi-task fitting and novel training losses. Extensive experiments on two challenging LPR datasets demonstrate the effectiveness of our proposed method in recovering the high-quality license plate image from the low-quality one and show that the the proposed method outperforms other state-of-the-art methods.

Young-min Song, Kwangjin Yoon, Young-Chul Yoon et al.

In this paper, we propose an efficient online multi-object tracking framework based on the GMPHD filter and occlusion group management scheme where the GMPHD filter utilizes hierarchical data association to reduce the false negatives caused by miss detection. The hierarchical data association consists of two steps: detection-to-track and track-to-track associations, which can recover the lost tracks and their switched IDs. In addition, the proposed framework is equipped with an object grouping management scheme which handles occlusion problems with two main parts. The first part is "track merging" which can merge the false positive tracks caused by false positive detections from occlusions, where the false positive tracks are usually occluded with a measure. The measure is the occlusion ratio between visual objects, sum-of-intersection-over-area (SIOA) we defined instead of the IOU metric. The second part is "occlusion group energy minimization (OGEM)" which prevents the occluded true positive tracks from false "track merging". We define each group of the occluded objects as an energy function and find an optimal hypothesis which makes the energy minimal. We evaluate the proposed tracker in benchmark datasets such as MOT15 and MOT17 which are built for multi-person tracking. An ablation study in training dataset shows that not only "track merging" and "OGEM" complement each other but also the proposed tracking method has more robust performance and less sensitive to parameters than baseline methods. Also, SIOA works better than IOU for various sizes of false positives. Experimental results show that the proposed tracker efficiently handles occlusion situations and achieves competitive performance compared to the state-of-the-art methods. Especially, our method shows the best multi-object tracking accuracy among the online and real-time executable methods.

Kwangjin Yoon, Young-min Song, Moongu Jeon

In this study, a multiple hypothesis tracking (MHT) algorithm for multi-target multi-camera tracking (MCT) with disjoint views is proposed. Our method forms track-hypothesis trees, and each branch of them represents a multi-camera track of a target that may move within a camera as well as move across cameras. Furthermore, multi-target tracking within a camera is performed simultaneously with the tree formation by manipulating a status of each track hypothesis. Each status represents three different stages of a multi-camera track: tracking, searching, and end-of-track. The tracking status means targets are tracked by a single camera tracker. In the searching status, the disappeared targets are examined if they reappear in other cameras. The end-of-track status does the target exited the camera network due to its lengthy invisibility. These three status assists MHT to form the track-hypothesis trees for multi-camera tracking. Furthermore, they present a gating technique for eliminating of unlikely observation-to-track association. In the experiments, they evaluate the proposed method using two datasets, DukeMTMC and NLPR-MCT, which demonstrates that the proposed method outperforms the state-of-the-art method in terms of improvement of the accuracy. In addition, they show that the proposed method can operate in real-time and online.

Young-chul Yoon, Abhijeet Boragule, Young-min Song et al.

In this paper, we propose the methods to handle temporal errors during multi-object tracking. Temporal error occurs when objects are occluded or noisy detections appear near the object. In those situations, tracking may fail and various errors like drift or ID-switching occur. It is hard to overcome temporal errors only by using motion and shape information. So, we propose the historical appearance matching method and joint-input siamese network which was trained by 2-step process. It can prevent tracking failures although objects are temporally occluded or last matching information is unreliable. We also provide useful technique to remove noisy detections effectively according to scene condition. Tracking performance, especially identity consistency, is highly improved by attaching our methods.