Sergiu Nedevschi

• PhD
• Professor at Technical University of Cluj-Napoca

Aerial scene understanding systems face stringent payload restrictions and must often rely on monocular depth estimation for modeling scene geometry, which is an inherently illposed problem. Moreover, obtaining accurate ground truth data required by learning-based methods raises significant additional challenges in the aerial domain. Self-supervise...

Landslides are global hazards that contribute significantly to worldwide catastrophes. Since landslides cause fatalities and property damage, understanding movement patterns is crucial to mitigate risks or potential reactivations. Slope kinematic modeling can utilize geodetic surveying or direct observations, though Unmanned Aerial Vehicles with ph...

The Hepatocellular Carcinoma (HCC) represents the most frequent malignant liver tumor. It evolves from cirrhosis after a restructuring phase, at the end of which dysplastic nodules result, which can transform into HCC. The needle biopsy is the golden standard for HCC diagnosis, being, however, invasive, dangerous, as it could lead to infections, re...

Large and high-quality training datasets are of critical importance for deep learning. In the context of the sematic segmentation challenge for UAV aerial images, we propose a strategy for data augmentation that can significantly reduce the effort of manually annotating a large number of images. The result is a set of semantic, depth and RGB images...

In the field of remote sensing, semantic segmentation of Unmanned Aerial Vehicle (UAV) imagery is crucial for tasks such as land resource management, urban planning, precision agriculture, and economic assessment. Traditional methods use Convolutional Neural Networks (CNNs) for hierarchical feature extraction but are limited by their local receptiv...

Cancer is one of the most severe diseases nowadays. Thus, tumor detection in a non-invasive and accurate manner is a challenging subject. Among these tumors, liver cancer is one of the most dangerous, being very common. Hepatocellular Carcinoma (HCC) is the most frequent malignant liver tumor. The golden standard for diagnosing HCC is mainly the bi...

The uncertainty that comes with planning, constructing, and maintaining buildings is a constant issue for architects and civil engineers. As topography is the framework that unites architecture and landscape, the design and planning projects heavily rely on a range of monitoring, surveying methods and comprehensive field data. Along with the tradit...

Hepatocellular Carcinoma (HCC) is the most frequent malignant liver tumor and the third cause of cancer-related deaths worldwide. For many years, the golden standard for HCC diagnosis has been the needle biopsy, which is invasive and carries risks. Computerized methods are due to achieve a noninvasive, accurate HCC detection process based on medica...

Monocular depth estimation (MDE) is one of the most difficult tasks in computer vision. The problem becomes even more complicated in case of aerial images due to the high complexity and lack of structure present in such scenarios. State of the art MDE methods can cope with such environments only by using high amounts of resources. In this work we t...

Depth-aware video panoptic segmentation tackles the inverse projection problem of restoring panoptic 3D point clouds from video sequences, where the 3D points are augmented with semantic classes and temporally consistent instance identifiers. We propose a novel solution with a multi-task network that performs monocular depth estimation and video pa...

We propose a novel solution for the task of video panoptic segmentation, that simultaneously predicts pixel-level semantic and instance segmentation and generates clip-level instance tracks. Our network, named VPS-Transformer, with a hybrid architecture based on the state-of-the-art panoptic segmentation network Panoptic-DeepLab, combines a convolu...

The European UP-Drive project addresses transportation-related challenges by providing key contributions that enable fully automated vehicle navigation and parking in complex urban areas, which results in a safer, inclusive, affordable and environmentally friendly transportation system. For this purpose, the project consortium developed a prototype...

Environment perception remains one of the key tasks in autonomous driving for which solutions have yet to reach maturity. Multi-modal approaches benefit from the complementary physical properties specific to each sensor technology used, boosting overall performance. The added complexity brought on by data fusion processes is not trivial to solve, w...

Panoptic segmentation provides a rich 2D environment representation by unifying semantic and instance segmentation. Most current state-of-the-art panoptic segmentation methods are built upon two-stage detectors and are not suitable for real-time applications, such as automated driving, due to their high computational complexity. In this work, we in...

Hepatocellular Carcinoma (HCC) is the most often met malignant liver tumor and one of the most frequent causes of death worldwide. The most reliable method for HCC diagnosis is the needle biopsy, but this is invasive, dangerous, leading to the spread of the malignity through the body, as well as to infections. We develop noninvasive, computerized t...

Object tracking is an essential problem in computer vision that has been extensively researched for decades. Tracking objects in thermal images is particularly difficult because of the lack of color information, low image resolution, or high similarity between objects of the same class. One of the main challenges in multi-object tracking, also refe...

Depth estimation approaches are crucial for environment perception in applications like autonomous driving or driving assistance systems. Solutions using cameras have always been preferred to other depth estimation methods, due to low sensor prices and their ability to extract rich semantic information from the scene. Monocular depth estimation alg...

In the context of a traffic scenario captured during night with infrared cameras we focus on pedestrian street cross action and we study the influence of the pedestrian pose with respect to the road environment on the accuracy of the action recognition model. This paper presents a complete frame work that performs pedestrian cross action recognitio...

Recent advances in 3D object detection focus on combining multiple 3D data representations in order to leverage each 3D representation. However, this opens a new issue: in- creased need for hardware memory and computational resources which are unfortunately limited. In this paper we explore a new 3D data representation, the Multi-Volume Grid repres...

The early recognition and understanding of the actions performed by pedestrians in traffic scenes leads to an anticipation of pedestrian intentions in advance and helps in the process of collision warning and avoidance in the context of autonomous vehicles. An environment with low visibility conditions such as night-time, fog, heavy rain or smoke i...

Hepatocellular Carcinoma (HCC) is the most common malignant liver tumor, being present in 70% of liver cancer cases. It usually evolves on the top of the cirrhotic parenchyma. The most reliable method for HCC diagnosis is the needle biopsy, which is an invasive, dangerous method. In our research, specific techniques for non-invasive, computerized H...

Environment perception by computing the depth is a key task for unmanned aerial vehicle (UAV) type systems. Due to the limited load they can carry, most drones are equipped with a single camera. This prevents general-purpose depth perception methods based either on light detection and ranging (LiDAR) or stereo reconstruction to be effectively used...

The identification of persons in multi-spectral images that are not spatially aligned is a challenging process. A correct identification can improve the pedestrian detection task for machine vision applications. In this context we propose a person identification mechanism able to correctly find the same person in infrared and visible images. The ma...

Hepatocellular carcinoma (HCC) is one of the leading causes of cancer-related deaths worldwide, with its mortality rate correlated with the tumor staging; i.e., early detection and treatment are important factors for the survival rate of patients. This paper presents the development of a novel visualization and detection system for HCC, which is a...

Drone perception systems use information from sensor fusion to perform tasks like object detection and tracking, visual localization and mapping, trajectory planning, and autonomous navigation. Applying these functions in real environments is a complex problem due to three-dimensional structures like trees, buildings, or bridges since the sensors (...

Hepatocellular Carcinoma (HCC) is the most frequent form of liver cancer, being the fourth leading cause of cancer-related death worldwide. The curative treatment in most cases is the tumor removal from the body (surgery), but more than 70% of HCC patients have advanced tumors and cannot be treated with such procedures. Alternative laparoscopic sur...

This paper presents our further efforts pertaining to the development of a noninvasive automated scoliosis screening and diagnostic solution, as well as some other spine disorders, using commercial-of-the-shelf (COTS) devices such as the recently announced Azure Kinect. The aim and main benefit of developing a MATLAB interface to the aforementioned...

The emergence of deep-learning methods in different computer vision tasks has proved to offer increased detection, recognition or segmentation accuracy when large annotated image datasets are available. In the case of medical image processing and computer-aided diagnosis within ultrasound images, where the amount of available annotated data is smal...

The stabilization and validation process of the measured position of objects is an important step for high-level perception functions and for the correct processing of sensory data. The goal of this process is to detect and handle inconsistencies between different sensor measurements, which result from the perception system. The aggregation of the...

In this paper we propose a robust curb detection method which is based on the fusion between semantically labeled camera images and a 3D point cloud coming from LiDAR sensors. The labels from the semantically enhanced cloud are used to reduce the curbs’ searching area. Several spatial cues are next computed on each candidate curb region. Based on t...

Deep learning requires large amounts of data for training models. For the task of semantic segmentation, manual annotation is time-consuming and difficult. With the recent advances in game engines, simulators have become more popular as they can instantly generate ground truth data for multiple sensors. In this paper, we make a thorough survey of t...

In this paper, we propose a novel semantic segmentation-based stereo reconstruction method that can keep up with the accuracy of the state-of-the art approaches while running in real time. The solution follows the classic stereo pipeline, each step in the stereo workflow being enhanced by additional information from semantic segmentation. Therefore...

Pole-like structures such as the ones used for traffic lights, traffic signs, utility poles, lampposts or even trees are encountered everywhere in urban scenarios. Because they are robust landmarks, they can help solve problems from the autonomous driving domain, such as localization, mapping, and navigation. In this paper, we propose a method that...

This article presents a new approach for detecting curbs in urban environments. It is based on the fusion between semantic labeled images obtained using a convolutional neural network and a LiDAR point cloud. Semantic information will be used in order to exploit context for the detection of urban curbs. Using only the semantic labels associated to...

In modern manufacturing plants, automation is widely adopted in the production phases, which leads to a high level of productivity and efficiency. However, the same level of automation is generally not achieved in logistics, typically performed by human operators and manually driven vehicles. In fact, even though automated guided vehicles (AGVs) ha...

This paper describes a super-sensor that enables 360-degree environment perception for automated vehicles in urban traffic scenarios. We use four fisheye cameras, four 360-degree LIDARs and a GPS/IMU sensor mounted on an automated vehicle to build a super-sensor that offers an enhanced low-level representation of the environment by harmonizing all...

This paper proposes a novel approach for segmenting and space partitioning data of sparse 3D LiDAR point clouds for autonomous driving tasks in urban environments. Our main focus is building a compact data representation which provides enough information for an accurate segmentation algorithm. We propose the use of an extension of elevation maps fo...