The ACM Lifelog Search Challenge (LSC) is an annual comparative benchmarking exercise that brings together researchers in the field of multimedia retrieval to evaluate interactive search systems using a large-scale multimodal lifelog dataset. This paper presents an overview of the ninth edition of the challenge (LSC’26), held as a workshop during the ACM International Conference on Multimedia Retrieval (ICMR ’26) in Amsterdam. To broaden its scope, the workshop now features three submission tracks: the traditional Challenge Track for real-time search performance, a new General Lifelog Research Track for theoretical and architectural advancements, and an additional Open Source Track aimed at enhancing reproducibility and reducing barriers to entry for new participants.

Eye gaze is fundamental to common sensorimotor activities. It directly indicates attention and its allocation, making it a critical human factor that reflects cognitive behavior. Data-driven measurements of gaze behavior quantitatively evaluate the allocation of visual attention, indicating the mental and physical activity status of individuals. Situational awareness provides a solid and semantically rich basis for gaze behavior during sensorimotor activities. However, few attempts have been made so far to measure eye gaze data from the perspective of situational awareness. In this paper, we propose four new computational measures of gaze behavior that align with the fundamental concept of situational awareness, effectively measuring the efficiency of visual scanning and sensorimotor activity. Our results have clearly shown that the proposed measures are effective and perform satisfactorily compared to the closely related methods. This work offers a new data-driven approach for biomarker discovery and behavioral biometrics using eye gaze data.

The coordination of eye and head in visual scanning is a fundamental behavior of humans in everyday sensorimotor activities such as walking and navigation. Deep understanding of the spatiotemporal dynamics of this coordination behavior undoubtedly plays an important role for many fields. However, relatively little is known about the computational and high-order interactions between eye and head in visual scanning during sensorimotor tasks. In this paper, based on the utilization of a recent tool from information theory, Partial Information Decomposition (PID), we quantify high-order components, namely uniqueness, redundancy, and synergy, in spatiotemporal interactions between eye movement and head motion time-series data during augmented reality-based navigation experience. To our knowledge, this is the first data-driven approach that leverages an information-theoretic tool to characterize high-order interactions involved in eye-head coordination during sensorimotor activities.

For the eighth time since 2018, the ACM Lifelog Search Challenge (LSC) was run to compare interactive lifelog search systems in a live metrics-based challenge. The goal of the LSC workshop is to comparatively evaluate the capabilities of systems accessing a large multimodal lifelog. LSC'25 attracted eleven participating teams, each of which had developed an innovative interactive lifelog retrieval system. The benchmark was organised in a hybrid manner (due to political issues in 2025) at the LSC workshop at ACM ICMR'25 in Chicago, USA. This short paper summarises the LSC workshop setting and presents the participating lifelog search systems.

Cataract surgery is the most frequently performed surgical procedure worldwide, involving the replacement of a patient's clouded eye lens with a synthetic intraocular lens to restore visual acuity. Although typically brief, the operation consists of distinct phases that demand precision and extensive training, traditionally constrained by the limitations of real-time observation under a microscope. To enhance learning and procedural accuracy, modern advancements in stereoscopic video capture and head-mounted displays (HMDs) offer a promising solution. This paper demonstrates the application of stereoscopic cataract surgery videos, visualized through Apple Vision Pro (AVP) and Meta Quest 3, to provide immersive 3D perspectives that enhance depth perception and spatial awareness. An expert evaluation study with experienced surgeons indicates that stereoscopic visualization significantly improves comprehension of spatial relationships and procedural maneuvers, suggesting its potential to revolutionize surgical education and real-time guidance in ophthalmic surgery.

To meet the growing demand for systematic surgical training, wet-lab environments have become indispensable platforms for hands-on practice in ophthalmology. Yet, traditional wet-lab training depends heavily on manual performance evaluations, which are labor-intensive, time-consuming, and often subject to variability. Recent advances in computer vision offer promising avenues for automated skill assessment, enhancing both the efficiency and objectivity of surgical education. Despite notable progress in ophthalmic surgical datasets, existing resources predominantly focus on real surgeries or isolated tasks, falling short of supporting comprehensive skill evaluation in controlled wet-lab settings. To address these limitations, we introduce WetCat, the first dataset of wet-lab cataract surgery videos specifically curated for automated skill assessment. WetCat comprises high-resolution recordings of surgeries performed by trainees on artificial eyes, featuring comprehensive phase annotations and semantic segmentations of key anatomical structures. These annotations are meticulously designed to facilitate skill assessment during the critical capsulorhexis and phacoemulsification phases, adhering to standardized surgical skill assessment frameworks. By focusing on these essential phases, WetCat enables the development of interpretable, AI-driven evaluation tools aligned with established clinical metrics. This dataset lays a strong foundation for advancing objective, scalable surgical education and sets a new benchmark for automated workflow analysis and skill assessment in ophthalmology training. The dataset and annotations are publicly available in Synapse (https://www.synapse.org/Synapse:syn66401174/files/).

Recent advances in deep learning have transformed computer-assisted intervention and surgical video analysis, driving improvements not only in surgical training, intraoperative decision support, and patient outcomes, but also in postoperative documentation and surgical discovery. Central to these developments is the availability of large, high-quality annotated datasets. In gynecologic laparoscopy, surgical scene understanding and action recognition are fundamental for building intelligent systems that assist surgeons during operations and provide deeper analysis after surgery. However, existing datasets are often limited by small scale, narrow task focus, or insufficiently detailed annotations, limiting their utility for comprehensive, end-to-end workflow analysis. To address these limitations, we introduce GynSurg, the largest and most diverse multi-task dataset for gynecologic laparoscopic surgery to date. GynSurg provides rich annotations across multiple tasks, supporting applications in action recognition, semantic segmentation, surgical documentation, and discovery of novel procedural insights. We demonstrate the dataset's quality and versatility by benchmarking state-of-the-art models under a standardized training protocol. To accelerate progress in the field, we publicly release the GynSurg dataset and its annotations (https://ftp.itec.aau.at/datasets/GynSurge/).

Egocentric video has seen increased interest in recent years, as it is used in a range of areas. However, most existing datasets are limited to a single perspective. In this paper, we present the CASTLE 2024 dataset, a multimodal collection containing ego- and exo-centric (i.e., first- and third-person perspective) video and audio from 15 time-aligned sources, as well as other sensor streams and auxiliary data. The dataset was recorded by volunteer participants over four days in a common location and includes the point of view of 10 participants, with an additional 5 fixed cameras providing an exocentric perspective. The entire dataset contains over 600 hours of UHD video recorded at 50 frames per second. In contrast to other datasets, CASTLE 2024 does not contain any partial censoring, such as blurred faces or distorted audio. The dataset is available via https://castle-dataset.github.io/.

The inaugural edition of the CASTLE grand challenge was held at ACM Multimedia 2025. The focus of the CASTLE challenge is to advance the state-of-the-art in analysis and understanding of multimodal data, especially centered around multistream ego- and exo-centric video. In this first instance of the challenge, participants had to solve three types of tasks - event instance search, object instance search, and question answering - in a fully automatic or interactive setting.

AR head-mounted displays (HMDs) facilitate pedestrian navigation by integrating AR guidance into users' field of view (FOV). Displaying AR guidance using a body-fixed coordinate system has the potential to further leverage this integration by enabling users to control when the guidance appears in their FOV. However, it remains unclear how to effectively position AR guidance within this coordinate system during pedestrian navigation. Therefore, we explored the effects of three AR guidance positions (top, middle, and bottom) within a body-fixed coordinate system on pedestrian navigation in a virtual environment. Our results showed that AR guidance position significantly influenced eye movements, walking behaviors, and subjective evaluations. The top position resulted in the shortest duration of fixations on the guidance compared to the middle and bottom positions, and lower mental demand than the bottom position. The middle position had the smallest rate of vertical eye movement during gaze shifts between the guidance and the environment, and the smallest relative difference in walking speed between fixations on the guidance and the environment compared to the top and bottom positions. The bottom position led to the shortest duration and smallest amplitude of gaze shifts between the guidance and the environment compared to the top and middle positions, and lower frustration than the top position. Based on these findings, we offer design implications for AR guidance positioning within a body-fixed coordinate system during pedestrian navigation.

The ability of contrastive language-image pretraining (CLIP) to pair text and image embeddings has notably enhanced the performance of video retrieval, enabling systems to efficiently utilize free-text searches within a CLIP-based embeddings index. Despite these technological improvements, disparities in user performance have been observed in recent video retrieval competitions, such as the Video Browser Showdown 2024 (VBS2024). We conclude that not only a highly accurate search index is important, but also factors such as runtime performance, system usability, browsing context, and interaction features. Addressing these issues, we introduce an enhanced version of the diveXplore system, merging the features of diveXplore 2024 and diveXB, our previously used systems for VBS2024 and IVR4B 2023. The system is designed for simple usage by video retrieval beginners, and optimized for fast and accurate search with clear content structuring and browsing context. The current diveXplore system uses an index of CLIP embeddings as the basic search method, including temporal queries, but also allows other search types (e.g., OCR, speech, metadata, etc.). It integrates a new video preview component with enriched browsing context, including a video summary of the related video, a video player, and a list of similar videos.

For the seventh time since 2018, the Lifelog Search Challenge (LSC) benchmarked interactive lifelog search systems in a live challenge. The LSC goal is to comparatively evaluate system capabilities to access large multimodal lifelogs comprising hundreds of thousands of records. LSC'24 attracted an unprecedented record number of twenty-one participating teams, where each team proposes innovative ideas implemented to new or already established interactive lifelog retrieval systems. The benchmark was organised in front of a live audience at the LSC workshop at ACM ICMR'24 in Phuket, Thailand. This short paper summarises the LSC workshop setting and presents the participating lifelog search systems.

Video streaming and its applications are growing rapidly, making video optimization a primary target for content providers looking to enhance their services. Enhancing the quality of videos requires the adjustment of different encoding parameters such as bitrate, resolution, and frame rate. To avoid brute force approaches for predicting optimal encoding parameters, video complexity features are typically extracted and utilized. To predict optimal encoding parameters effectively, content providers traditionally use unsupervised feature extraction methods, such as ITU-T’s Spatial Information (SI) and Temporal Information (TI) to represent the spatial and temporal complexity of video sequences. Recently, Video Complexity Analyzer (VCA) was introduced to extract DCT-based features to represent the complexity of a video sequence (or parts thereof). These unsupervised features, however, cannot accurately predict video encoding parameters. To address this issue, this paper introduces a novel supervised feature extraction method named DeepVCA, which extracts the spatial and temporal complexity of video sequences using deep neural networks. In this approach, the encoding bits required to encode each frame in intra-mode and inter-mode are used as labels for spatial and temporal complexity, respectively. Initially, we benchmark various deep neural network structures to predict spatial complexity. We then leverage the similarity of features used to predict the spatial complexity of the current frame and its previous frame to rapidly predict temporal complexity. This approach is particularly useful as the temporal complexity may depend not only on the differences between two consecutive frames but also on their spatial complexity. Our proposed approach demonstrates significant improvement over unsupervised methods, especially for temporal complexity. As an example application, we verify the effectiveness of these features in predicting the encoding bitrate and encoding time of video sequences, which are crucial tasks in video streaming. The source code and dataset is available at https://github.com/cd-athena/ DeepVCA.

Adaptive live video streaming applications utilize a predefined collection of bitrate-resolution pairs, known as abitrate ladder, for simplicity and efficiency, eliminating the need for additional run-time to determine the optimal pairs during the live streaming session. These applications do not incorporate two-pass encoding methods due to increased latency. However, an optimized bitrate ladder could result in lower storage and delivery costs and improvedQuality of Experience(QoE). This paper presents a Just Noticeable Difference (JND)-aware constrained Variable Bitrate (cVBR) Two-pass Per-title encoding Scheme (JTPS) designed specifically for live video streaming. JTPS predicts a content- and JND-aware bitrate ladder using low-complexity features based onDiscrete Cosine Transform(DCT) energy and optimizes the constant rate factor (CRF) for each representation using random forest-based models. The effectiveness of JTPS is demonstrated using the open source video encoder x265, with an average bitrate reduction of 18.80% and 32.59% for the same PSNR and VMAF, respectively, compared to the standardHTTP Live Streaming(HLS) bitrate ladder using Constant Bitrate (CBR) encoding. The implementation of JTPS also resulted in a 68.96% reduction in storage space and an 18.58% reduction in encoding time for a JND of six VMAF points.

We present an interactive video retrieval system called diveXB (diveXplore for Beginners), which is an enhancement of our diveXplore system used for the recent Video Browser Showdown competitions. In addition to a common free-text search with text-image embeddings, diveXB integrates a new exploration component that allows browsing the video collection without any query. For this purpose, we employ k-means clustering of image embeddings that are extracted from video summaries with state-of-the-art vision transformers and select the most coherent clusters for query-free browsing in our user interface. First qualitative experiments have shown that this feature allows users to easily get an overview of the content structure in the video collection and to find desired content without the need for any textual input.

Action recognition is a prerequisite for many applications in laparoscopic video analysis including but not limited to surgical training, operation room planning, follow-up surgery preparation, post-operative surgical assessment, and surgical outcome estimation. However, automatic action recognition in laparoscopic surgeries involves numerous challenges such as (I) cross-action and intra-action duration variation, (II) relevant content distortion due to smoke, blood accumulation, fast camera motions, organ movements, object occlusion, and (III) surgical scene variations due to different illuminations and viewpoints. Besides, action annotations in laparoscopy surgeries are limited and expensive due to requiring expert knowledge. In this study, we design and evaluate a CNN-RNN architecture as well as a customized training-inference framework to deal with the mentioned challenges in laparoscopic surgery action recognition. Using stacked recurrent layers, our proposed network takes advantage of inter-frame dependencies to negate the negative effect of content distortion and variation in action recognition. Furthermore, our proposed frame sampling strategy effectively manages the duration variations in surgical actions to enable action recognition with high temporal resolution. Our extensive experiments confirm the superiority of our proposed method in action recognition compared to static CNNs.

For the sixth time since 2018, the Lifelog Search Challenge (LSC) was organized as a comparative benchmarking exercise for various interactive lifelog search systems. The goal of this international competition is to test system capabilities to access large multimodal lifelogs. LSC’23 attracted twelve participanting teams, each of whom had developed a competitive interactive lifelog retrieval system. The benchmark was organized in front of live audience at the LSC workshop at ACM ICMR’23. As in previous editions, this introductory paper presents the LSC workshop and introduces the participating lifelog search systems.

With the emergence of multiple modern video codecs, streaming service providers are forced to encode, store, and transmit bitrate ladders of multiple codecs separately, consequently suffering from additional energy costs for encoding, storage, and transmission. To tackle this issue, we introduce an online energy-efficient Multi-Codec Bitrate ladder Estimation scheme (MCBE) for adaptive video streaming applications. In MCBE, quality representations within the bitrate ladder of new-generation codecs (e.g., High Efficiency Video Coding (HEVC), Alliance for Open Media Video 1 (AV1)) that lie below the predicted rate-distortion curve of the Advanced Video Coding (AVC) codec are removed. Moreover, perceptual redundancy between representations of the bitrate ladders of the considered codecs is also minimized based on a Just Noticeable Difference (JND) threshold. Therefore, random forest-based models predict the VMAF score of bitrate ladder representations of each codec. In a live streaming session where all clients support the decoding of AVC, HEVC, and AV1, MCBE achieves impressive results, reducing cumulative encoding energy by 56.45%, storage energy usage by 94.99%, and transmission energy usage by 77.61% (considering a JND of six VMAF points). These energy reductions are in comparison to a baseline bitrate ladder encoding based on current industry practice.

In recent years, the utilization intensity and thus the demand for storing cataract surgery videos for different purposes has increased. Hospitals continuously improve their technical recording equipment, i.e., cameras, to enhance the post-operative processing efficiency of the recordings. However, afterward, the videos are stored on hospitals' internal data servers in their original size, which leads to a massive storage consumption. In this paper, we propose a relevance-based compression scheme. First, we perform a user study with clinicians to define the relevance rates of regular cataract surgery phases. Then, we compress different phases based on the determined relevance rates, using different encoding parameters and two coding standards, namely H.264/AVC and AV1. Afterward, the medical experts evaluate the visual quality of the encoded videos. Our results show a storage-saving potential for H.264/AVC of up to 95.94% and up to 98.82% for AV1, excluding idle phases (no tools are visible).

The creation of datasets in medical imaging is a central topic of research, especially with the advances of deep learning in the past decade. Publications of such datasets typically report baseline results with one or more deep neural networks in the form of established performance metrics (e.g., F1-score, Jaccard, etc.). Then, much work is done trying to beat these baseline metrics to compare different neural architectures. However, these reported metrics are almost meaningless when the underlying data does not conform to specific standards. In order to better understand what standards we need, we have reproduced and analyzed a study of four medical image classification datasets in laparoscopy. With automated frame extraction of surgical videos, we find that the resulting images are way too similar and produce high evaluation metrics by design. We show this similarity with a basic SIFT algorithm that produces high evaluation metrics on the original data. We confirm our hypothesis by creating and evaluating a video-based dataset split from the original images. The original network evaluated on the video-based split performs worse than our basic SIFT algorithm on the original data.

For the fifth time since 2018, the Lifelog Search Challenge (LSC) facilitated a benchmarking exercise to compare interactive search systems designed for multimodal lifelogs. LSC'22 attracted nine participating research groups who developed interactive lifelog retrieval systems enabling fast and effective access to lifelogs. The systems competed in front of a hybrid audience at the LSC workshop at ACM ICMR'22. This paper presents an introduction to the LSC workshop, the new (larger) dataset used in the competition, and introduces the participating lifelog search systems.

During the last 10 years of Video Browser Showdown (VBS), there were many different approaches tested for known-item search and ad-hoc search tasks. Undoubtedly, teams incorporating state-of-the-art models from the machine learning domain had an advantage over teams focusing just on interactive interfaces. On the other hand, VBS results indicate that effective means of interaction with a search system is still necessary to accomplish challenging search tasks. In this tutorial, we summarize successful deep models tested at the Video Browser Showdown as well as interfaces designed on top of corresponding distance/similarity spaces. Our broad experience with competition organization and evaluation will be presented as well, focusing on promising findings and also challenging problems from the most recent iterations of the Video Browser Showdown.

The retrieval of multimedia content remains a difficult problem where a high accuracy or specificity can often only be achieved interactively, with a user working closely and iteratively with a retrieval system. While there exist several venues for the exchange of insights in the area of information retrieval in general and multimedia retrieval specifically, there is little discussion on such interactive retrieval approaches. The Workshop on Interactive Multimedia Retrieval offers such a venue. Held for the 2nd time in 2022, it attracted a diverse set of contributions, six of which were accepted for presentation. The following provides a brief overview of the workshop itself as well as the contributions of 2022.

Abstract Endometriosis is a common gynecologic condition typically treated via laparoscopic surgery. Its visual versatility makes it hard to identify for non-specialized physicians and challenging to classify or localize via computer-aided analysis. In this work, we take a first step in the direction of localized endometriosis recognition in laparoscopic gynecology videos using region-based deep neural networks Faster R-CNN and Mask R-CNN. We in particular use and further develop publicly available data for transfer learning deep detection models according to distinctive visual lesion characteristics. Subsequently, we evaluate the performance impact of different data augmentation techniques, including selected geometrical and visual transformations, specular reflection removal as well as region tracking across video frames. Finally, particular attention is given to creating reasonable data segmentation for training, validation and testing. The best performing result surprisingly is achieved by randomly applying simple cropping combined with rotation, resulting in a mean average segmentation precision of 32.4% at 50-95% intersection over union overlap (64.2% for 50% overlap).

Endometriosis is a common women's condition exhibiting a manifold visual appearance in various body-internal locations. Having such properties makes its identification very difficult and error-prone, at least for laymen and non-specialized medical practitioners. In an attempt to provide assistance to gynecologic physicians treating endometriosis, this demo paper describes a system that is trained to segment one frequently occurring visual appearance of endometriosis, namely dark endometrial implants. The system is capable of analyzing laparoscopic surgery videos, annotating identified implant regions with multi-colored overlays and displaying a detection summary for improved video browsing.

Nowadays, modern ski resorts provide additional services to customers, such as recording videos of specific moments from their skiing experience. This and similar tasks can be achieved by using computer vision methods. In this work, we evaluate the detection performance of current object detection methods and the tracking performance of a detection-based tracking algorithm. The evaluation is based on videos of skiers and snowboarders from ski resorts. We collect videos of race tracks from different resorts and compile a public dataset of images and videos, where skiers and snowboarders are annotated with bounding boxes. Based on this data, we evaluate the performance of four state-of-the-art object detection methods. This evaluation is performed with general models trained on the MS COCO dataset as well as with custom models trained on our dataset. In addition, we review the performance of the detection-based, multi-object tracking algorithm Deep SORT, which we adapt for skier tracking.The results show promising performance and reveal that the MS COCO models already achieve high Precision, while training a custom model additionally improves the performance. Bigger models profit from custom training in terms of more accurate bounding box placement and higher Precision, while smaller models have an overall high training payoff. The modified Deep SORT tracker manages to follow a skier's trajectory over an extended period and operates with high accuracy, which indicates that the tracker is overall well suited for tracking of skiers and snowboarders on race tracks. Even when exposed to strong camera and skier movement changes, the tracker stays latched onto the target.

Visual scanning plays an important role in sampling visual information from the surrounding environments for a lot of everyday sensorimotor tasks, such as driving. In this paper, we consider the problem of visual scanning mechanism underpinning sensorimotor tasks in 3D dynamic environments. We exploit the use of eye tracking data as a behaviometric, for indicating the visuo-motor behavioral measure in the context of virtual driving. A new metric of visual scanning efficiency (VSE), which is defined as a mathematical divergence between a fixation distribution and a distribution of optical flows induced by fixations, is proposed by making use of a widely-known information theoretic tool, namely the square root of Jensen-Shannon divergence. Psychophysical eye tracking studies, in virtual reality based driving, are conducted to reveal that the new metric of visual scanning efficiency can be employed very well as a proxy evaluation for driving performance. These results suggest that the exploitation of eye tracking data provides an effective behaviometric for sensorimotor activities.

The Lifelog Search Challenge (LSC) is an annual benchmarking challenge for comparing approaches to interactive retrieval from multi-modal lifelogs. LSC'21, the fourth challenge, attracted sixteen participants, each of which had developed interactive retrieval systems for large multimodal lifelogs. These interactive retrieval systems participated in a comparative evaluation in front of an online live-audience at the LSC workshop at ACM ICMR'21. This overview presents the motivation for LSC'21, the lifelog dataset used in the competition, and the participating systems.

Intraoperative tracking of laparoscopic instruments is often a prerequisite for computer and robotic-assisted interventions. While numerous methods for detecting, segmenting and tracking of medical instruments based on endoscopic video images have been proposed in the literature, key limitations remain to be addressed: Firstly, robustness, that is, the reliable performance of state-of-the-art methods when run on challenging images (e.g. in the presence of blood, smoke or motion artifacts). Secondly, generalization; algorithms trained for a specific intervention in a specific hospital should generalize to other interventions or institutions. In an effort to promote solutions for these limitations, we organized the Robust Medical Instrument Segmentation (ROBUST-MIS) challenge as an international benchmarking competition with a specific focus on the robustness and generalization capabilities of algorithms. For the first time in the field of endoscopic image processing, our challenge included a task on binary segmentation and also addressed multi-instance detection and segmentation. The challenge was based on a surgical data set comprising 10,040 annotated images acquired from a total of 30 surgical procedures from three different types of surgery. The validation of the competing methods for the three tasks (binary segmentation, multi-instance detection and multi-instance segmentation) was performed in three different stages with an increasing domain gap between the training and the test data. The results confirm the initial hypothesis, namely that algorithm performance degrades with an increasing domain gap. While the average detection and segmentation quality of the best-performing algorithms is high, future research should concentrate on detection and segmentation of small, crossing, moving and transparent instrument(s) (parts).

Abstract Despite all its irrefutable benefits, the development of steganography methods has sparked ever-increasing concerns over steganography abuse in recent decades. To prevent the inimical usage of steganography, steganalysis approaches have been introduced. Since motion vector manipulation leads to random and indirect changes in the statistics of videos, MV-based video steganography has been the center of attention in recent years. In this paper, we propose a 54-dimentional feature set exploiting spatio-temporal features of motion vectors to blindly detect MV-based stego videos. The idea behind the proposed features originates from two facts. First, there are strong dependencies among neighboring MVs due to utilizing rate-distortion optimization techniques and belonging to the same rigid object or static background. Accordingly, MV manipulation can leave important clues on the differences between each MV and the MVs belonging to the neighboring blocks. Second, a majority of MVs in original videos are locally optimal after decoding concerning the Lagrangian multiplier, notwithstanding the information loss during compression. Motion vector alteration during information embedding can affect these statistics that can be utilized for steganalysis. Experimental results have shown that our features’ performance far exceeds that of state-of-the-art steganalysis methods. This outstanding performance lies in the utilization of complementary spatio-temporal statistics affected by MV manipulation as well as feature dimensionality reduction applied to prevent overfitting. Moreover, unlike other existing MV-based steganalysis methods, our proposed features can be adjusted to various settings of the state-of-the-art video codec standards such as sub-pixel motion estimation and variable-block-size motion estimation.

Despite the fact that automatic content analysis has made remarkable progress over the last decade - mainly due to significant advances in machine learning - interactive video retrieval is still a very challenging problem, with an increasing relevance in practical applications. The Video Browser Showdown (VBS) is an annual evaluation competition that pushes the limits of interactive video retrieval with state-of-the-art tools, tasks, data, and evaluation metrics. In this paper, we analyse the results and outcome of the 8th iteration of the VBS in detail. We first give an overview of the novel and considerably larger V3C1 dataset and the tasks that were performed during VBS 2019. We then go on to describe the search systems of the six international teams in terms of features and performance. And finally, we perform an in-depth analysis of the per-team success ratio and relate this to the search strategies that were applied, the most popular features, and problems that were experienced. A large part of this analysis was conducted based on logs that were collected during the competition itself. This analysis gives further insights into the typical search behavior and differences between expert and novice users. Our evaluation shows that textual search and content browsing are the most important aspects in terms of logged user interactions. Furthermore, we observe a trend towards deep learning based features, especially in the form of labels generated by artificial neural networks. But nevertheless, for some tasks, very specific content-based search features are still being used. We expect these findings to contribute to future improvements of interactive video search systems.

Comprehensive and fair performance evaluation of information retrieval systems represents an essential task for the current information age. Whereas Cranfield-based evaluations with benchmark datasets support development of retrieval models, significant evaluation efforts are required also for user-oriented systems that try to boost performance with an interactive search approach. This article presents findings from the 9th Video Browser Showdown, a competition that focuses on a legitimate comparison of interactive search systems designed for challenging known-item search tasks over a large video collection. During previous installments of the competition, the interactive nature of participating systems was a key feature to satisfy known-item search needs, and this article continues to support this hypothesis. Despite the fact that top-performing systems integrate the most recent deep learning models into their retrieval process, interactive searching remains a necessary component of successful strategies for known-item search tasks. Alongside the description of competition settings, evaluated tasks, participating teams, and overall results, this article presents a detailed analysis of query logs collected by the top three performing systems, SOMHunter, VIRET, and vitrivr. The analysis provides a quantitative insight to the observed performance of the systems and constitutes a new baseline methodology for future events. The results reveal that the top two systems mostly relied on temporal queries before a correct frame was identified. An interaction log analysis complements the result log findings and points to the importance of result set and video browsing approaches. Finally, various outlooks are discussed in order to improve the Video Browser Showdown challenge in the future.

Automatically detecting surgical tools in recorded surgery videos is an important building block of further content-based video analysis. In ophthalmology, the results of such methods can support training and teaching of operation techniques and enable investigation of medical research questions on a dataset of recorded surgery videos. While previous methods used frame-based classification techniques to predict the presence of surgical tools - but did not localize them, we apply a recent deep-learning segmentation method (Mask R-CNN) to localize and segment surgical tools used in ophthalmic cataract surgery. We add ground-truth annotations for multi-class instance segmentation to two existing datasets of cataract surgery videos and make resulting datasets publicly available for research purposes. In the absence of comparable results from literature, we tune and evaluate the Mask R-CNN approach on these datasets for instrument segmentation/localization and achieve promising results (61% mean average precision on 50% intersection over union for instance segmentation, working even better for bounding box detection or binary segmentation), establishing a reasonable baseline for further research. Moreover, we experiment with common data augmentation techniques and analyze the achieved segmentation performance with respect to each class (instrument), providing evidence for future improvements of this approach.

In cataract surgery, the operation is performed with the help of a microscope. Since the microscope enables watching real-time surgery by up to two people only, a major part of surgical training is conducted using the recorded videos. To optimize the training procedure with the video content, the surgeons require an automatic relevance detection approach. In addition to relevance-based retrieval, these results can be further used for skill assessment and irregularity detection in cataract surgery videos. In this paper, a three-module framework is proposed to detect and classify the relevant phase segments in cataract videos. Taking advantage of an idle frame recognition network, the video is divided into idle and action segments. To boost the performance in relevance detection, the cornea where the relevant surgical actions are conducted is detected in all frames using Mask R-CNN. The spatiotemporally localized segments containing higher-resolution information about the pupil texture and actions, and complementary temporal information from the same phase are fed into the relevance detection module. This module consists of four parallel recurrent CNNs being responsible to detect four relevant phases that have been defined with medical experts. The results will then be integrated to classify the action phases as irrelevant or one of four relevant phases. Experimental results reveal that the proposed approach outperforms static CNNs and different configurations of feature-based and end-to-end recurrent networks.

The Lifelog Search Challenge (LSC) is an annual comparative benchmarking activity for comparing approaches to interactive retrieval from multi-modal lifelogs. LSC'20, the third such challenge, attracts fourteen participants with their interactive lifelog retrieval systems. These systems are comparatively evaluated in front of a live-audience at the LSC workshop at ACM ICMR'20 in Dublin, Ireland. This overview motivates the challenge, presents the dataset and system configuration used in the challenge, and briefly presents the participating teams.

Accumulating recordings of daily conducted surgical interventions such as endoscopic procedures for the long term generates very large video archives that are both difficult to search and explore. Since physicians utilize this kind of media routinely for documentation, treatment planning or education and training, it can be considered a crucial task to make said archives manageable in regards to discovering or retrieving relevant content. We present an interactive tool including a multitude of modalities for browsing, searching and filtering medical content, demonstrating its usefulness on over 140 hours of pre-processed laparoscopic surgery videos.

Recorded cataract surgery videos play a prominent role in training and investigating the surgery, and enhancing the surgical outcomes. Due to storage limitations in hospitals, however, the recorded cataract surgeries are deleted after a short time and this precious source of information cannot be fully utilized. Lowering the quality to reduce the required storage space is not advisable since the degraded visual quality results in the loss of relevant information that limits the usage of these videos. To address this problem, we propose a relevance-based compression technique consisting of two modules: (i) relevance detection, which uses neural networks for semantic segmentation and classification of the videos to detect relevant spatio-temporal information, and (ii) content-adaptive compression, which restricts the amount of distortion applied to the relevant content while allocating less bitrate to irrelevant content. The proposed relevance-based compression framework is implemented considering five scenarios based on the definition of relevant information from the target audience's perspective. Experimental results demonstrate the capability of the proposed approach in relevance detection. We further show that the proposed approach can achieve high compression efficiency by abstracting substantial redundant information while retaining the high quality of the relevant content.

The Video Browser Showdown (VBS) has influenced the Multimedia community already for 10 years now. More than 30 unique teams from over 21 countries participated in the VBS since 2012 already. In 2021, we are celebrating the 10th anniversary of VBS, where 17 international teams compete against each other in an unprecedented contest of fast and accurate multimedia retrieval. In this tutorial we discuss the motivation and details of the VBS contest, including its history, rules, evaluation metrics, and achievements for multimedia retrieval. We talk about the properties of specific VBS retrieval systems and their unique characteristics, as well as existing open-source tools that can be used as a starting point for participating for the first time. Participants of this tutorial get a detailed understanding of the VBS and its search systems, and see the latest developments of interactive video retrieval.

Recorded videos from surgeries have become an increasingly important information source for the field of medical endoscopy, since the recorded footage shows every single detail of the surgery. However, while video recording is straightforward these days, automatic content indexing - the basis for content-based search in a medical video archive - is still a great challenge due to the very special video content. In this work, we investigate segmentation and recognition of surgical instruments in videos recorded from laparoscopic gynecology. More precisely, we evaluate the achievable performance of segmenting surgical instruments from their background by using a region-based fully convolutional network for instance-aware (1) instrument segmentation as well as (2) instrument recognition. While the first part addresses only binary segmentation of instances (i.e., distinguishing between instrument or background) we also investigate multi-class instrument recognition (i.e., identifying the type of instrument). Our evaluation results show that even with a moderately low number of training examples, we are able to localize and segment instrument regions with a pretty high accuracy. However, the results also reveal that determining the particular instrument is still very challenging, due to the inherently high similarity of surgical instruments.

The Video Browser Showdown (VBS) is a challenging video retrieval competition annually performed at the International Conference on MultiMedia Modeling (MMM). This paper presents a review of all eight years of the VBS performed so far - as a summary of the keynote presented at CBMI 2019. It discusses the different types of tasks and sessions, provides an overview of the datasets and participating teams, and it gives some insights on the setup and evaluation procedure. It also evaluates the progress made over the last few years in terms of average task solve time and task solve ratio.

In this work we analyze content statistics of the V3C1 dataset, which is the first partition of theVimeo Creative Commons Collection (V3C). The dataset has been designed to represent true web videos in the wild, with good visual quality and diverse content characteristics, and will serve as evaluation basis for the Video Browser Showdown 2019-2021 and TREC Video Retrieval (TRECVID) Ad-Hoc Video Search tasks 2019-2021. The dataset comes with a shot segmentation (around 1 million shots) for which we analyze content specifics and statistics. Our research shows that the content of V3C1 is very diverse, has no predominant characteristics and provides a low self-similarity. Thus it is very well suited for video retrieval evaluations as well as for participants of TRECVID AVS or the VBS.

We present a tutorial focusing on video retrieval tasks, where state-of-the-art deep learning approaches still benefit from interactive decisions of users. The tutorial covers general introduction to the interactive video retrieval research area, state-of-the-art video retrieval systems, evaluation campaigns and recently observed results. Moreover, a significant part of the tutorial is dedicated to a practical exercise with three selected state-of-the-art systems in the form of an interactive video retrieval competition. Participants of this tutorial will gain a practical experience and also a general insight of the interactive video retrieval topic, which is a good start to focus their research on unsolved challenges in this area.

In recent years, we have observed a rise of interest in the multimedia community towards research topics related to health. It can be observed that this goes into two interesting directions. One is personal health with a larger focus on well-being and everyday healthy living. The other direction focuses more on multimedia challenges within the health-care systems, for example, how can multimedia content produced in hospitals be used efficiently but also on the user perspective of patients and health-care personal. Challenges and requirements in this interesting and challenging direction are similar to classic multimedia research, but with some additional pitfalls and challenges. This tutorial aims to give a general introduction to the research area; to provide an overview of specific requirements, pitfalls and challenges; to discuss existing and possible future work; and to elaborate on how machine learning approaches can help in multimedia-related challenges to improve the health-care quality for patients and support medical experts in their daily work.

Computers are powerful tools capable of solving a great variety of ever so complex problems, yet training them to interpret even the simplest video scenes can prove more challenging than one might imagine. Still being one of the major problems in computer vision, this issue recently is addressed by utilizing promising deep learning approaches in order to recognize objects and their semantics. For achieving this goal, huge artificial networks are fed with many human-created annotations using more or less sophisticated tools for speeding up the otherwise time-consuming task of manual annotation. Purposefully refraining from designing yet another of these annotation tools, in this work we strive for evaluating what makes existing ones great or not, i.e. we aim at determining effectiveness and efficiency of state-of-the-art object annotation tools when employed for annotating different kinds of video content. Our findings in a user study evaluating three comparable tools on three videos of distinct domains indicate a significant difference in annotation effort from a video perspective, yet no significance regarding utilized tools. Further, we determine a significant correlation between annotation time and accuracy.

This work summarizes the findings of the 7th iteration of the Video Browser Showdown (VBS) competition organized as a workshop at the 24th International Conference on Multimedia Modeling in Bangkok. The competition focuses on video retrieval scenarios in which the searched scenes were either previously observed or described by another person (i.e., an example shot is not available). During the event, nine teams competed with their video retrieval tools in providing access to a shared video collection with 600 hours of video content. Evaluation objectives, rules, scoring, tasks, and all participating tools are described in the article. In addition, we provide some insights into how the different teams interacted with their video browsers, which was made possible by a novel interaction logging mechanism introduced for this iteration of the VBS. The results collected at the VBS evaluation server confirm that searching for one particular scene in the collection when given a limited time is still a challenging task for many of the approaches that were showcased during the event. Given only a short textual description, finding the correct scene is even harder. In ad hoc search with multiple relevant scenes, the tools were mostly able to find at least one scene, whereas recall was the issue for many teams. The logs also reveal that even though recent exciting advances in machine learning narrow the classical semantic gap problem, user-centric interfaces are still required to mediate access to specific content. Finally, open challenges and lessons learned are presented for future VBS events.

The most essential step towards semiautomatic extraction of relevant surgery scenes is semantic understanding of surgical actions in surgery videos. Currently, Convolutional Neural Networks (CNNs) are a de-facto standard for automatic content classification in many domain, including medical imaging. We aim to include increase the predictive performance of surgical action recognition within gynecologic laparoscopy, a subfield of endoscopic surgery, by fusing temporal information to the input layer of CNNs (early fusion), as well as temporal aggregation of single-frame prediction results (late fusion). Our evaluation shows that the proposed early fusion approaches are able to outperform a single-frame baseline when using the GoogLeNet architecture. Moreover, early fusion of motion information benefits the classification performance regardless of late fusion strategy. Late fusion has a high impact on classification performance, and its increase is additive to the performance increase of early fusion. Eventually, we found that the CNN capacity influences these results drastically. We conclude that the proposed methods in combination with a sufficiently high CNN capacity allow for a substantial increase in predictive performance.q

Modern day endoscopic technology enables medical staff to conveniently document surgeries via recording raw treatment footage, which can be utilized for planning further proceedings, future case revisitations or even educational purposes. However, the prospect of manually perusing recorded media files constitutes a tedious additional workload on physicians' already packed timetables and therefore ultimately represents a burden rather than a benefit. The aim of this PhD project is to improve upon this situation by closely collaborating with medical experts in order to devise datasets and systems to facilitate semi-automatic post-surgical media processing.

Multimedia research is becoming more and more important for the medical domain, where an increasing number of videos and images are integrated in the daily routine of surgical and diagnostic work. While the collection of medical multimedia data is not an issue, appropriate tools for efficient use of this data are missing. This includes management and inspection of the data, visual analytics, as well as learning relevant semantics and using recognition results for optimizing surgical and diagnostic processes. The characteristics and requirements in this interesting but challenging field are different than the ones in classic multimedia domains. Therefore, this tutorial gives a general introduction to the field, provides a broad overview of specific requirements and challenges, discusses existing work and open challenges, and elaborates in detail how machine learning approaches can help in multimedia-related fields to improve the performance of surgeons/clinicians.

In this tutorial we discuss interactive video search tools and methods, review their need in the age of deep learning, and explore video and multimedia search challenges and their role as evaluation benchmarks in the field of multimedia information retrieval. We cover three different campaigns (TRECVID, Video Browser Showdown, and the Lifelog Search Challenge), discuss their goals and rules, and present their achieved findings over the last half-decade. Moreover, we talk about datasets, tasks, evaluation procedures, and examples of interactive video search tools, as well as how they evolved over the years. Participants of this tutorial will be able to gain collective insights from all three challenges and use them for focusing their research efforts on outstanding problems that still remain unsolved in this area.

Modern imaging technology enables medical practitioners to perform minimally invasive surgery (MIS), i.e. a variety of medical interventions inflicting minimal trauma upon patients, hence, greatly improving their recoveries. Not only patients but also surgeons can benefit from this technology, as recorded media can be utilized for speeding-up tedious and time-consuming tasks such as treatment planning or case documentation. In order to improve the predominantly manually conducted process of analyzing said media, with this work we publish four datasets extracted from gynecologic, laparoscopic interventions with the intend on encouraging research in the field of post-surgical automatic media analysis. These datasets are designed with the following use cases in mind: medical image retrieval based on a query image, detection of instrument counts, surgical actions and anatomical structures, as well as distinguishing on which anatomical structure a certain action is performed. Furthermore, we provide suggestions for evaluation metrics and first baseline experiments.

Cataract surgery is one of the most frequently performed microscopic surgeries in the field of ophthalmology. The goal behind this kind of surgery is to replace the human eye lense with an artificial one, an intervention that is often required due to aging. The entire surgery is performed under microscopy, but co-mounted cameras allow to record and archive the procedure. Currently, the recorded videos are used in a postoperative manner for documentation and training. An additional benefit of recording cataract videos is that they enable video analytics (i.e., manual and/or automatic video content analysis) to investigate medically relevant research questions (e.g., the cause of complications). This, however, necessitates a medical multimedia information system trained and evaluated on existing data, which is currently not publicly available. In this work we provide a public video dataset of 101 cataract surgeries that were performed by four different surgeons over a period of 9 months. These surgeons are grouped into moderately experienced and highly experienced surgeons (assistant vs. senior physicians), providing the basis for experience-based video analytics. All videos have been annotated with quasi-standardized operation phases by a senior ophthalmic surgeon.

In recent years, digital endoscopy has established as key technology for medical screenings and minimally invasive surgery. Since then, various research communities with manifold backgrounds have picked up on the idea of processing and automatically analyzing the inherently available video signal that is produced by the endoscopic camera. Proposed works mainly include image processing techniques, pattern recognition, machine learning methods and Computer Vision algorithms. While most contributions deal with real-time assistance at procedure time, the post-procedural processing of recorded videos is still in its infancy. Many post-processing problems are based on typical Multimedia methods like indexing, retrieval, summarization and video interaction, but have only been sparsely addressed so far for this domain. The goals of this survey are (1) to introduce this research field to a broader audience in the Multimedia community to stimulate further research, (2) to describe domain-specific characteristics of endoscopic videos that need to be addressed in a pre-processing step, and (3) to systematically bring together the very diverse research results for the first time to provide a broader overview of related research that is currently not perceived as belonging together.

Videos of endoscopic surgery are used for education of medical experts, analysis in medical research, and documentation for everyday clinical life. Hand-crafted image descriptors lack the capabilities of a semantic classification of surgical actions and video shots of anatomical structures. In this work, we investigate how well single-frame convolutional neural networks (CNN) for semantic shot classification in gynecologic surgery work. Together with medical experts, we manually annotate hours of raw endoscopic gynecologic surgery videos showing endometriosis treatment and myoma resection of over 100 patients. The cleaned ground truth dataset comprises 9 h of annotated video material (from 111 different recordings). We use the well-known CNN architectures AlexNet and GoogLeNet and train these architectures for both, surgical actions and anatomy, from scratch. Furthermore, we extract high-level features from AlexNet with weights from a pre-trained model from the Caffe model zoo and feed them to an SVM classifier. Our evaluation shows that we reach an average recall of .697 and .515 for classification of anatomical structures and surgical actions respectively using off-the-shelf CNN features. Using GoogLeNet, we achieve a mean recall of .782 and .617 for classification of anatomical structures and surgical actions respectively. With AlexNet the achieved recall is .615 for anatomical structures and .469 for surgical action classification respectively. The main conclusion of our work is that advances in general image classification methods transfer to the domain of endoscopic surgery videos in gynecology. This is relevant as this domain is different from natural images, e.g. it is distinguished by smoke, reflections, or a limited amount of colors.

With a rigorous long-term archival of endoscopic surgeries, vast amounts of video and image data accumulate. Surgeons are not able to spend their valuable time to manually search within endoscopic multimedia databases (EMDBs) or manually maintain links to interesting sections in order to quickly retrieve relevant surgery sections. Enabling the surgeons to quickly access the relevant surgery scenes, we utilize the fact that surgeons record external images additionally to the surgery video and aim to link them to the appropriate video sequence in the EMDB using a query-by-example approach. We propose binary Convolutional Neural Network (CNN) features off-the-shelf and compare them to several baselines: pixel-based comparison (PSNR), image structure comparison (SSIM), hand-crafted global features (CEDD and feature signatures), as well as CNN baselines Histograms of Class Confidences (HoCC) and Neural Codes (NC). For evaluation, we use 5.5 h of endoscopic video material and 69 query images selected by medical experts and compare the performance of the aforementioned image mathing methods in terms of video hit rate and distance to the true playback time stamp (PTS) for correct video predictions. Our evaluation shows that binary CNN features are compact, yet powerful image descriptors for retrieval in the endoscopic imaging domain. They are able to maintain state-of-the-art performance, while providing the benefit of low storage space requirements and hence provide the best compromise.

In the domain of gynecologic surgery an increasing number of surgeries are performed in a minimally invasive manner. These laparoscopic surgeries require specific psychomotor skills of the operating surgeon, which are difficult to learn and teach. This is the reason why an increasing number of surgeons promote checking video recordings of laparoscopic surgeries for the occurrence of technical errors with surgical actions. This manual surgical quality assessment (SQA) process, however, is very cumbersome and time-consuming when carried out without any support from content-based video retrieval. In this work we propose a video content descriptor called MIDD (Motion Intensity and Direction Descriptor) that can be effectively used to find similar segments in a laparoscopic video database and thereby help surgeons to more quickly inspect other instances of a given error scene. We evaluate the retrieval performance of MIDD with surgical actions from gynecologic surgery in direct comparison to several other dynamic content descriptors. We show that the MIDD descriptor significantly outperforms the state-of-the-art in terms of retrieval performance as well as in terms of runtime performance. Additionally, we release the manually created video dataset of 16 classes of surgical actions from medical laparoscopy to the public, for further evaluations.

The last decade has seen innovations that make video recording, manipulation, storage, and sharing easier than ever before, thus impacting many areas of life. New video retrieval scenarios emerged as well, which challenge the state-of-the-art video retrieval approaches. Despite recent advances in content analysis, video retrieval can still benefit from involving the human user in the loop. We present our experience with a class of interactive video retrieval scenarios and our methodology to stimulate the evolution of new interactive video retrieval approaches. More specifically, the video browser showdown evaluation campaign is thoroughly analyzed, focusing on the years 2015-2017. Evaluation scenarios, objectives, and metrics are presented, complemented by the results of the annual evaluations. The results reveal promising interactive video retrieval techniques adopted by the most successful tools and confirm assumptions about the different complexity of various types of interactive retrieval scenarios. A comparison of the interactive retrieval tools with automatic approaches (including fully automatic and manual query formulation) participating in the TRECVID 2016 ad hoc video search task is discussed. Finally, based on the results of data analysis, a substantial revision of the evaluation methodology for the following years of the video browser showdown is provided.

Due to increasing possibilities to create digital video, we are facing the emergence of large video archives that are made accessible either online or offline. Though a lot of research has been spent on video retrieval tools and methods, which allow for automatic search in videos, still the performance of automatic video retrieval is far from optimal. At the same time, the organization of personal data is receiving increasing research attention due to the challenges that are faced in gathering, enriching, searching and visualizing this data. Given the increasing quantities of personal data being gathered by individuals, the concept of a heterogeneous personal digital libraries of rich multimedia and sensory content for every individual is becoming a reality.

When regarding physicians' tremendously packed timetables, it comes as no surprise that they start managing even critical situations hastily in order to cope with the high demands laid out for them. Apart from treating patients' conditions they as well are required to perform time-consuming administrative tasks, including post-surgery video analyses. Concerning documentation of minimally invasive surgeries (MIS), specifically endoscopy, such processes usually involve repeatedly perusing through lengthy, in the worst case uncut recordings - a redundant task that nowadays can be optimized by using readily available technology: we present a tool for annotating endoscopic video frames targeting a specific use case - endometriosis, i.e. the dislocation of uterine-like tissue.

The domain of minimally invasive surgery has recently attracted attention from the Multimedia community due to the fact that systematic video documentation is on the rise in this medical field. The vastly growing volumes of video archives demand for effective and efficient techniques to retrieve specific information from large video collections with visually very homogeneous content. One specific challenge in this context is to retrieve scenes showing similar surgical actions, i.e., similarity search. Although this task has a high and constantly growing relevance for surgeons and other health professionals, it has rarely been investigated in the literature so far for this particular domain. In this paper, we propose and evaluate a number of both static and dynamic content descriptors for this purpose. The former only take into account individual images, while the latter consider the motion within a scene. Our experimental results show that although static descriptors achieve the highest overall performance, dynamic descriptors are much more discriminative for certain classes of surgical actions. We conclude that the two approaches have complementary strengths and further research should investigate methods to combine them.

The rapidly increasing volume of videos recorded in the course of endoscopic screenings and surgeries poses demanding challenges to video retrieval and browsing systems. Surgeons typically have to use standard video players to retrospectively review their procedures, which is an extremely cumbersome and time-consuming process. We present an HTML5-based video explorer that is specially tailored to this purpose and enables a time-efficient post-operative review of procedures. It incorporates various interactive browsing mechanisms as well as domain-specific content-based features based on previous research results. Preliminary interviews with surgeons indicate that this tool can considerably improve retrieval and browsing efficiency for users in the medical domain and allows surgeons to more easily and quickly revisit specific moments in recordings of their endoscopic surgeries.

In hospitals all around the world, medical multimedia information systems have gained high importance over the last few years. One of the reasons is that an increasing number of interventions are performed in a minimally invasive way. These endoscopic inspections and surgeries are performed with a tiny camera -- the endoscope -- which produces a video signal that is used to control the intervention. Apart from the viewing purpose, the video signal is also used for automatic content analysis during the intervention as well as for post-surgical usage, such as communicating operation techniques, planning future interventions, and medical forensics. Another reason is video documentation, which is even enforced by law in some countries. The problem, however, is the sheer amount of unstructured medical videos that are added to the multimedia archive on a daily basis. Without proper management and a multimedia information system, the medical videos cannot be used efficiently for post-surgical scenarios. It is therefore already foreseeable that medical multimedia information systems will gain even more attraction in the next few years. In this tutorial we will introduce the audience to this challenging new field, describe the domain-specific characteristics and challenges of medical multimedia data, introduce related use cases, and talk about existing works -- contributed by the medical imaging and robotics community, but also already partly from the multimedia community -- as well as the many open issues and challenges that bear high research potential.

Interactive video retrieval tools developed over the past few years are emerging as powerful alternatives to automatic retrieval approaches by giving the user more control as well as more responsibilities. Current research tries to identify the best combinations of image, audio and text features that combined with innovative UI design maximize the tools performance. We present the last installment of the Video Browser Showdown 2015 which was held in conjunction with the International Conference on MultiMedia Modeling 2015 (MMM 2015) and has the stated aim of pushing for a better integration of the user into the search process. The setup of the competition including the used dataset and the presented tasks as well as the participating tools will be introduced . The performance of those tools will be thoroughly presented and analyzed. Interesting highlights will be marked and some predictions regarding the research focus within the field for the near future will be made.

Common lossy video compression methods have already reached a very high performance level and are about to reach their natural limit. However, their compression efficiency is still insufficient for certain domains where a plethora of video data should be archived. One example is the domain of medical endoscopy where entire surgeries are recorded for documentation with high visual quality because it is not possible to anticipate if and when exceptional situations may happen. Eventually, only a fraction of the video footage is of actual relevance for post-operative review. In this paper, we introduce the concept of domain-specific video compression by the example of this very specific domain. We identify several characteristics of endoscopic surgery videos that can be exploited by content-based analysis algorithms to achieve a significantly higher compression efficiency than in current practice. The extensive evaluation shows that the proposed methods massively reduce the data volume without any semantic information loss and thus allow for long-term video archiving of endoscopic surgery videos. Moreover, we expect to stimulate future research in this challenging new field.

We present a browser for endoscopic videos that is designed to easily navigate and compare scenes on a tablet. It utilizes frame stripes of different levels of detail to quickly switch between fast and detailed navigation. Moreover, it uses saliency methods to determine which areas of a given keyframe contain the most information to further improve the visualization of the frame stripes. As scenes with much movement can be non-relevant out-of-patient scenes, the tool supports filtering for scenes of low, medium and high motion. The tool can be especially useful for patient debriefings as well as for educational purposes.

We present a tool for mobile browsing and annotation tailored for endoscopic videos. Professional users can utilize this tablet app for patients debriefings or educational purposes. It supports text input, free-hand and shape drawing as well as angle measurements, e.g. for comparing instrument orientation. It is possible to annotate single frames as well as user-defined video sections. Moreover, it provides easy and efficient navigation via a zoom-able navigation bar that is based on frame stripes. Frame stripes are created by extracting a single, one pixel wide vertical stripe from every keyframe. The stripes are then arranged next to each other to form a uniform bar. This gives users great overview of the content of a given video. Furthermore, the app supports creation of custom reports based on the entered annotations that can be directly mailed or printed for further usage.

We present a new approach for collaborative video search and video browsing relying on a combination of traditional, index-based video retrieval complemented with large-scale human-based visual inspection. In particular, a traditional PC interface is used for query-based search using advanced indexing and querying methods (e.g., concept search), whereas a visualization of the database on a tablet is used for pure human-based browsing. Both parts are coupled to compensate for mutual disadvantages; human visual inspection allows for a better, more detailed analysis of the data - also bridging the semantic gap - but comes at the price of an unfiltered database - a disadvantage that is resolved by using the results from the query engine to change visualization order.

We present a novel content-based video retrieval tool that facilitates interactive search in large video archives by focusing on the factors content context and content dynamics. It incorporates query-by-concept and query-by-sketch functionality. For the latter we introduce temporal feature signatures - an extension to color feature signatures by adding the dimension of content dynamics over time. Moreover, temporal feature signatures are also used for performing segment similarity searches for improved matching performance in contrast to static solutions. Because of intelligent caching this process is performed in just a couple of seconds, which improves the overall user experience greatly. Found segments can also be easily displayed in their chronological context for refining the search. To better visualize the content dynamics of a video segment users just need to move their mouse cursor across a segments thumbnail to peek into its content.

In the field of medical endoscopy more and more surgeons are changing over to record and store videos of their endoscopic procedures, such as surgeries and examinations, in long-term video archives. In order to support surgeons in accessing these endoscopic video archives in a content-based way, we propose a simple yet effective signature-based approach: the Signature Matching Distance based on adaptive-binning feature signatures. The proposed distance-based similarity model facilitates an adaptive representation of the visual properties of endoscopic images and allows for matching these properties efficiently. We conduct an extensive performance analysis with respect to the task of linking specific endoscopic images with video segments and show the high efficacy of our approach. We are able to link more than 88% of the endoscopic images to their corresponding correct video segments, which improves the current state of the art by one order of magnitude.

We propose a new interface that facilitates content navigation in videos on devices with touchscreen interaction. This interface allows both coarse-grained and fine-grained navigation in an intuitive way and enables better performance when used to locate specific scenes in videos. We implemented this interface on a 5.5-inch smartphone and tested it with 24 users. Our results show that for video navigation tasks the proposed interface significantly outperforms the seekerbar interface, commonly used with video players on mobile devices. Moreover, we found that the interaction concept of the Scrubbing Wheel has a much lower perceived workload than the widely used seeker-bar, and is the preferred tool to locate scenes in videos for all tested users in our study.

With an increasing amount of video data in our daily life, the need for content-based search in videos increases as well. Though a lot of research has been spent on video retrieval tools and methods which allow for automatic search in videos through content-based queries, still the performance of automatic video retrieval is far from optimal. In this tutorial we discussed (i) proposed solutions for improved video content navigation, (ii) typical interaction of content-based querying features, and (iii) advanced video content visualization methods. Moreover, we discussed interactive video search systems and ways to evaluate their performance.

In recent years it has become common practice to archive video recordings of laparoscopic surgeries for documentation purposes and for retrospective review. Typically, the videos are captured in High Definition (HD) format but encoded with legacy coding standards like MPEG-2 requiring an enormous storage capacity. In this paper we present the results of a subjective quality assessment study with 37 medical experts. We identify appropriate encoding configurations for the H.264 AVC coding standard to guarantee a visually loss less quality with a significant bitrate reduction. Further, we show that it is not necessary to capture the highest possible quality for documentation and retrospective analysis. A lower technical quality with a substantially lower bitrate still provides sufficient semantic quality. We finally present basic recommendations for an efficient encoding strategy with an appropriate tradeoff between visual quality and bitrate.

This paper proposes a shot-based key frame selection technique, aiming at generating a condensed set of frames representing the essential content of a video sequence. Inspired by the successful utilization of Jensen-Shannon Divergence (JSD) and Jensen-Rényi Divergence (JRD) for key frame selection [1] [2], we investigate several popularly accepted f-divergences for calculating the frame-by-frame distance to segment the video clip and then to obtain the key frames. Based on simulation and real test videos, the performances of the key frame selection method by using different versions of f-divergences are systematically analyzed. Extensive experimentation shows that, compared with the methods using JSD and JRD, the new technique is slightly better and computationally faster.

Mobile devices like smartphones and tablets are becoming increasingly capable in terms of processing power. Although they are already used in computer vision, no comparable measurement experiments of the popular OpenCV framework have been made yet. We try to fill this gap by evaluating the performance of a set of typical OpenCV operations, on mobile devices like the iPad Air and iPhone 5S. We compare those results with the performance of a consumer grade laptop PC (MacBook Pro). Our tests span from simple image manipulation methods to keypoint detection and descriptor extraction as well as descriptor matching. Results show that the top performing device can match the performance of the PC up to 80 percent in specific operations.

We propose a new interface for browsing large image collections on touch-enabled devices, which allows for faster visual search than a common grid interface. The proposed interface consists of 3D rings arranged in a vertical stack. Each ring can be rotated independently with a horizontal wipe gesture and the whole stack can be moved up and down with vertical wipe gestures. We present results from a user study with 16 participants, where we compared the performance of the proposed interface to the one provided by the default image browsing interface on tablets. The evaluation results show that the proposed interface performs significantly faster than the default interface on the tablet and achieves a significantly better rating by the users.

Most interfaces in the field of image and video search use a two-dimensional grid interface, which presents image thumbnails in a left-to-right arrangement that can be browsed from top to bottom. This grid interface, however, has several drawbacks that become particularly apparent when performing interactive search tasks for target items in large collections of images or videos. Therefore, we propose to use 3-D interfaces as an alternative to the grid interface for interactive known-item search in visual data as they can partially overcome these drawbacks. In this paper, we first summarize our ideas and discuss design aspects of a 3-D ring and a 3-D globe interface. Next, we present results from four different user studies, where we evaluated the performance of these interfaces for known-item search tasks in image collections. Our results from these studies show that the proposed 3-D interfaces allow for significantly faster visual target search on desktop computers with mouse interaction as well as on tablet devices. The interfaces also achieve better subjective ratings. However, our evaluation also shows that on smartphones with 3.5-in screens an improvement over the grid interface in terms of visual search time is only possible in collections with more than 200 images.

The actual content of endoscopic videos is typically limited to a circular area in the image center. This area has a dynamic position and size and is surrounded by a dark, but noisy border. In this paper we present a novel algorithm that (1) classifies which frames of an endoscopic video feature the circular content area and (2) determines its exact position and size, if present. This information is very useful for improving the performance of subsequent analysis techniques. It can also be used for more efficient video encoding and economic printing of still images in findings and reports. The evaluation shows that the proposed method is very accurate, robust and efficient in terms of runtime.

Smartphones and tablets are popular devices. As lightweight, compact devices with built-in high-quality cameras, they are ideal to carry around and to use for snapshot photography. As the number of photos accumulate on the device quickly finding a particular photo can be tedious using the default grid-based photo browser installed on the device. In this paper we investigate user performance in a photo browsing task on an iPad and an iPod Touch device. We present results from two user experiments comparing the standard grid interface to a pan-and-zoom able grid, a 3D-globe and a 3D-ring. In particular we are interested in how the interfaces perform with large photo collections (100 to 400 photos). The results show most promise for the pan-and-zoom grid and that the performance with the standard grid interface quickly deteriorates with large collections.

In recent years, it became common to record video footage of laparoscopic surgeries. This leads to large video archives that are very hard to manage. They often contain a considerable portion of completely irrelevant scenes which waste storage capacity and hamper an efficient retrieval of relevant scenes. In this paper we (1) define three classes of irrelevant segments, (2) propose visual feature extraction methods to obtain irrelevance indicators for each class and (3) present an extensible framework to detect irrelevant segments in laparoscopic videos. The framework includes a training component that learns a prediction model using nonlinear regression with a generalized logistic function and a segment composition algorithm that derives segment boundaries from the fuzzy frame classifications. The experimental results show that our method performs very good both for the classification of individual frames and the detection of segment boundaries in videos and enables considerable storage space savings.

As mobile devices become more and more pervasive, they are increasingly used to record and watch personal and professional videos. Common video browsers on smart phones and tablets, however, fail to provide users an efficient and engaging experience to browse through video content. In this demo paper, we present an early prototype for browsing videos on tablets: browsing video with a 3D filmstrip. A video is split up into equidistant, uniformly sampled time segments. The segments are represented by key-frames shown on the surface of a filmstrip, allowing an immediate overview of great parts of the video. Further, a user can scroll through the filmstrip, start playback for any segment and refine the preview of segments by simple touch gestures.

In this work we propose an early prototype of a video browser for mobile devices with touchscreens. We concentrate on utilizing the thumbs because of the natural posture used with the devices when watching videos in landscape mode. The controls are only displayed when the user touches the screen and automatically rearrange themselves depending on the position of the thumbs. A combination of a radial menu and an extended seeker control with hierarchical browsing and bookmarking features enables the user to navigate quickly through videos.

Interactive image and video search tools typically use a grid-like arrangement of thumbnails for preview purpose. Such a display, which is commonly known as storyboard, provides limited flexibility at interactive search and it does not optimally exploit the available screen estate. In this paper we design and evaluate alternatives to the common two-dimensional storyboard. We take advantage of 3D graphics in order to present image thumbnails in cylindrical arrangements. Through a user study we evaluate the performance of these interfaces in terms of visual search time and subjective performance.

With users increasingly using their mobile devices such as smartphones as digital photo albums, effective methods for managing these collections are becoming increasingly important. Standard solutions provide only limited facilities for organising, browsing and searching image collections on mobile devices, making it challenging and time-consuming to locate images of interest.

Default image browsing interfaces on touch-based mobile devices provide limited support for image search tasks. To facilitate fast and convenient searches we propose an alternative interface that takes advantage of 3D graphics and arranges images on a rotatable globe according to color similarity. In a user study we compare the new design to the iPad's image browser. Results collected from 24 participants show that for color-sorted image collections the globe can reduce search time by 23% without causing more errors and that it is perceived as being fun to use and preferred over the standard browsing interface by 70% of the participants.

We investigate whether users' visual search performance in a commonly used grid-like arrangement of images (i.e., a storyboard) can be improved by using a similarity-based sorting of images. We propose a simple but efficient algorithm for sorting image based on their color similarity. The algorithm generates an intuitive arrangement of images and allows for general application with several different layouts (e.g., storyboard, simple row/column, 3D globe/cylinder). In difference to previous work, which rarely present results from user studies, we perform a fair user study and compare an interface with color sorted images to an interface with images positioned in a random order. Both interfaces use exactly the same screen estate and interaction means. Results show that users are 20% faster with the sorted interface.

We present an intuitive user interface for the exploration of images on mobile multi-touch devices. Our interface uses a novel cylindrical 3D visualization of visually sorted images as well as touch gestures and tilting operations to support mobile users in interactive browsing of images by providing convenient navigation/interaction and intuitive visualization capabilities.

We demonstrate an interactive 3D storyboard that take advantage of 3D graphics in order to overcome certain limitations of conventional 2D storyboards when used for the task of image and video browsing.

We present a video browsing tool that combines advantages of the hierarchical browsing concept with 3D projection and multi-threaded programming in order to provide a convenient and efficient interface. The tool allows for instantaneous hierarchical browsing of video and uses a dynamic approach (i.e. tree of playable video segments instead of static key frames)that also supports parallel playback.

We present a tool that can efficiently facilitate interactive navigation and search in videos. In addition to browsing a video by shots it also allows a user to navigate through a video with extended seeker bars showing time-related content abstractions. Users having a rough knowledge about the content characteristics of scenes to be found can efficiently use these extended seeker bars to quickly find these scenes by interactive navigation. Moreover, users can easily perform similarity queries by utilizing content knowledge that has been gained during the browsing/navigation process. These queries can also be stored and reused for search in other videos having similar/same content. Furthermore, the tool can execute many queries at once and visualize the results as semantic events in a different seeker bar. Our tool provides a real alternative for situations where a user currently needs to employ a common video player for the task of search and navigation.

We propose a video browsing tool supporting new efficient navigation means and content-based search within a single video, allowing for interactive exploration and playback of video content. The user interface provides flexible navigation indices by visualizing low-level features and frame surrogates along one or more timelines, called interactive navigation summaries. By applying simple and fast content analysis, navigation summary computation becomes feasible during browsing, enabling addition, removal, and update of navigation summaries at runtime. Semantically similar video segments will be visualized by similar patterns in certain navigation summaries, which enables users to quickly recognize and navigate to potential similar segments. Moreover, free-shape regions of video frames or video segments within navigation summaries can be selected by the user to launch a fast content-based search to find frames with similar regions or segments with similar navigation summaries.

We present a video browsing tool that uses a novel and powerful visualization technique of video motion. The tool provides an interactive navigation index that allows users to quickly and easily recognize content semantics like scenes with fast/slow motion (in general or according to a specific direction), scenes showing still/moving objects in front of a still/moving background, camera pans, or camera zooms. Moreover, the visualization facilitates identification of similar segments in a video. A first user study has shown encouraging results.

We present a new approach for video browsing using visualization of motion direction and motion intensity statistics by color and brightness variations. Statistics are collected from motion vectors of H.264/AVC encoded video streams, so full video decoding is not required. By interpreting visualized motion patterns of video segments, users are able to quickly identify scenes similar to a prototype scene or identify potential scenes of interest. We give some examples of motion patterns with different semantic value, including camera zooms, hill jumps of ski-jumpers, and the repeated appearance of a news speaker. In a user study we show that certain scenes of interest can be found significantly faster using our video browsing tool than using a video player with VCR-like controls.

We present a video content browsing tool which is supposed to be used for two purposes: (1) efficiently searching a certain scene within a long video sequence and (2) quickly identifying videos out of interest. Instead of a complex user interface with many features we rather use a simple but flexible player-like interface which can, however, support the user on taking advantage of his/her knowledge about the content semantics. The content analysis is based on some simple features which can be extracted from compressed data of H264/AVC and, thus, allow very fast analysis. An early user study has shown encouraging results.