RGB-D cameras, such as the Microsoft Kinect, provide us with the 3D information, color and depth, associated with the scene. Interactive 3D Tele-Immersion (i3DTI) systems use such RGB-D cameras to capture the person present in the scene in order to collaborate with other remote users and interact with the virtual objects present in the environment. Using a single camera, it becomes difficult to estimate an accurate skeletal pose and complete 3D model of the person, especially when the person is not in the complete view of the camera. With multiple cameras, even with partial views, it is possible to get a more accurate estimate of the skeleton of the person leading to a better and complete 3D model. In this paper, we present a real-time skeletal pose identification approach that leverages on the inaccurate skeletons of the individual Kinects, and provides a combined optimized skeleton. We estimate the Probability of an Accurate Joint (PAJ) for each joint from all of the Kinect skeletons. We determine the correct direction of the person and assign the correct joint sides for each skeleton. We then use a greedy consensus approach to combine the highly probable and accurate joints to estimate the combined skeleton. Using the individual skeletons, we segment the point clouds from all the cameras. We use the already computed PAJ values to obtain the Probability of an Accurate Bone (PAB). The individual point clouds are then combined one segment after another using the calculated PAB values. The generated combined point cloud is a complete and accurate 3D representation of the person present in the scene. We validate our estimated skeleton against two well-known methods by computing the error distance between the best view Kinect skeleton and the estimated skeleton. An exhaustive analysis is performed by using around 500000 skeletal frames in total, captured using 7 users and 7 cameras. Visual analysis is performed by checking whether the estimated skeleton is completely present within the human model. We also develop a 3D Holo-Bubble game to showcase the real-time performance of the combined skeleton and point cloud. Our results show that our method performs better than the state-of-the-art approaches that use multiple Kinects, in terms of objective error, visual quality and real-time user performance.

Applications involving 3D scanning and reconstruction & 3D Tele-immersion provide an immersive experience by capturing a scene using multiple RGB-D cameras, such as Kinect. Prior knowledge of intrinsic calibration of each of the cameras, and extrinsic calibration between cameras, is essential to reconstruct the captured data. The intrinsic calibration for a given camera rarely ever changes, so only needs to be estimated once. However, the extrinsic calibration between cameras can change, even with a small nudge to the camera. Calibration accuracy depends on sensor noise, features used, sampling method, etc., resulting in the need for iterative calibration to achieve good calibration.

With the rising popularity of handheld virtual reality (VR) devices and depth sensing RGB-D cameras, a variety of VR applications merging these two technologies has been suggested. However, immersive quality of experience in such VR applications is constrained mainly by the large data size and the hardware limitations to handle it. The depth data captured by RGB-D cameras provide a dense sampling of the surface, resulting in a high-poly mesh, which is difficult to be rendered on handheld VR devices due to their limited processing power. To improve the immersive VR experience, a sparse approximation of the depth data is needed. Traditional mesh and point cloud simplification methods are iterative and so are unsuitable for real-time applications. In this paper, we introduce a depth-imagebased approach that is capable of generating a good quality sparse mesh for visualization in real time. We propose a curvaturesensitive surface simplification-CS3operator that assigns an importance measure to each point in the depth image, based on the local curvature. Further, it applies an importance-order-based restrictive sampling to generate a sparse representation that retains the overall shape as well as the finer features of the object. We also modify the 2-D sweep-line-based constrained Delaunay triangulation to generate 3-D meshes from the sparse point sampling obtained using CS3. In addition, the proposed approach preserves key surface properties, such as texture coordinates and materials. We used three different datasets containing dense 3-D models with and without texture, which are scanned using various sensors to validate and compare the robustness, real-time performance, and accuracy of the proposed method over existing approaches. Based on the experimental results, we show that the proposed CS3operator and modified 2-D sweep-line-based triangulation generate sparse meshes from depth image in real time, performing significantly faster than current state-of-the-art methods while maintaining similar visual quality.

Users' Quality of Experience (QoE) in Interactive 3D Tele-Immersion (i3DTI) systems is influenced by several factors such as the quality of the "live" 3D avatars of the users, network latency, rendering methodology (head mounted display or regular TV type of display), etc. Hence, it becomes important to answer the question: "Is Visual Quality (VQ) the only factor to be considered or do better immersion and faster interactions matter, in having good QoE?" To answer this question, in this paper, a highly optimized state-of-the-art i3DTI framework implementation is introduced along with a soccer-inspired penalty shootout game. This game allows users to experience various situations, view different angles, perceive delays, track virtual ball motion, and play naturally using their entire bodies. A head mounted display device - Oculus Rift allows the users to get completely immersed and perform better in the penalty shootout game compared to watching themselves play on a 3D TV. This scenario is obtained in a controlled lab setting with ultra-high-speed network that has ultra-low latency, high VQ, fast and realistic interactions. Such ideal conditions are not typically available in a wide area Internet. Hence, for faster interaction scenario, we used lesser RGB-D cameras for 3D reconstruction, thereby reducing the model quality significantly. The high responsiveness of the game masked the user's perception of quality, and resulted in them not noticing the lower VQ of the reconstructed 3D models. Based on the results from the user study that focused on immersion, interaction and VQ aspects; users felt the game to be visually appealing, intuitive, engaging, and highly entertaining in all of the scenarios.

The current state-of-the-art tele-medicine applications only allow audiovisual communication between a doctor and the patient, necessitating a clinician to physically examine the patient. The doctor relies on the physical examination performed by the clinician, along with the audiovisual dialogue with the patient. In this paper, a Haptic-enabled Tele-Immersive Musculoskeletal Examination (H-TIME) system is introduced, that allows doctors to physically examine musculoskeletal conditions of the patients remotely, by looking at the 3D reconstructed model of the patient in the virtual world, and physically feeling the patient's range of mobility using a haptic device. The proposed bidirectional haptic rendering in H-TIME can allow the doctor to evaluate a patient who suffers from problems in their upper extremities, such as the shoulder, elbow, wrist, etc., and evaluate them remotely. Real world user study was performed, between the doctors and the patients, and it highlighted the potential of the proposed system. The study indicated a high degree of correlation between the in-person and H-TIME evaluations of the patient. Both the doctors and patients involved in the study, felt that the system could potentially replace in-person consultations, someday.

3D Tele-Immersion systems allow geographically distributed users to interact in a virtual world using their "live" 3D models. The capture, reconstruction, transfer, and rendering of these models introduce significant latency into the system. Implicit Latency (ℒ') can be estimated using system clocks to measure the time after the data was received from the RGB-D camera, till the request to render the result. The Observed Latency (ℒ) between a real world event and the event being rendered on the display, cannot be accurately represented by ℒ' since ℒ' ignores the time taken to capture, or update the display, etc. In this paper, a Visual Pattern based Latency Estimation (VPLE) approach is introduced to calculate the real world visual latency of a system without the need for any custom hardware. VPLE generates a constantly changing pattern that is captured and rendered by the 3DTI system. An external observer records both the pattern and the rendered results at high frame rates. ℒ is estimated by calculating the difference between the generated and rendered patterns. VPLE is extended to allow ℒ estimation between geographically distributed sites. Evaluations show that the accuracy of VPLE depends on the refresh rate of the pattern, and is within 4ms. ℒ of a distributed 3DTI system implemented on the GPU is significantly lower than the CPU implementation, and is comparable to video streaming. It is also shown that the ℒ' estimates for GPU based 3DTI implementations are off by almost 100% compared to the ℒ.

RGB-D cameras have enabled real-time 3D video processing for numerous computer vision applications, especially for surveillance type applications. In this paper, we first present a real-time anti-forensic 3D object stream manipulation framework to capture and manipulate live RBG-D data streams to create realistic images/videos showing individuals performing activities they did not actually do. The framework uses computer vision and graphics methods to render photorealistic animations of live mesh models captured using the camera. Next, we conducted a visual inspection of the manipulated RGB-D streams (just like security personnel would do) by users who are computer vision and graphics scientists. The study shows that it was significantly difficult to distinguish between the real or reconstructed rendering of such 3D video sequences, thus clearly showing the potential security risk involved. Finally, we investigate the efficacy of forensic approaches for detecting such manipulations.

3D Tele-Immersion (3DTI) has emerged as an efficient environment for virtual interactions and collaborations in a variety of fields like rehabilitation, education, gaming, etc. In 3DTI, geographically distributed users are captured using multiple cameras and immersed in a single virtual environment. The quality of experience depends on the available network bandwidth, quality of the 3D model generated and the time taken for rendering. In a collaborative environment, achieving high quality, high frame rate rendering by transmitting data to multiple sites having different bandwidth is challenging. In this paper we introduce a network adaptive textured mesh generation scheme to transmit varying quality data based on the available bandwidth. To reduce the volume of information transmitted, a visual quality based vertex selection approach is used to generate a sparse representation of the user. This sparse representation is then transmitted to the receiver side where a sweep-line based technique is used to generate a 3D mesh of the user. High visual quality is maintained by transmitting a high resolution texture image compressed using a lossy compression algorithm. In our studies users were unable to notice visual quality variations of the rendered 3D model even at 90% compression.

Augmented Reality (AR) and Augmented Virtuality (AV) systems have been used in various fields such as entertainment, broadcasting, gaming [1], etc. Collaborative AR or AV (CAR/CAV) systems are a special kind of such system in which the interaction happens through the exchange of multi-modal data between multiple users/sites. Multiple sensors capture the real objects and enable interaction with shared virtual objects in a customizable virtual environment. Haptic devices can be added to introduce force feedback when the virtual objects are manipulated. These applications are demanding in terms of network resources to support low latency media delivery and media source switching similar to broadcast applications. Enabling real time interaction with multiple modalities with high volume data requires an advanced media transport protocol that supports low latency media delivery and fast media source (channel) switching. To enable such collaboration over a stochastic network like the Internet requires a combination of technologies from data design, synchronization to real time media delivery. MPEG Media Transport (MMT) [ISO/IEC 23008-1] is a new standard suite of protocols designed to work with demanding, real-time interactive multimedia applications, typically in the context of one-to-one and one-to-many communication. In this paper, we identify the augmentations that are required for the many-to-many nature of CAR/CAV applications and propose MMT+AVR as a middle ware solution for use in CAV applications. Through an example CAV application implemented on top of MMT+AVR, we show how it provides efficient support for developing CAV applications with ease.

3D Tele-Immersion (3DTI) systems capture and transmit large volumes of data per frame to enable virtual world interaction between geographically distributed people. Large delays/latencies introduced during the transmission of these large volumes of data can lead to poor quality of experience of the 3DTI systems. Such poor experiences can possibly be overcome by animating the previously received mesh using the current skeletal data (that is very small in size and hence experiences much lower communication delays). However, using just the previously transmitted mesh for animation is not ideal and could render inconsistent results. In this paper, we present a DIstortion Score based Pose SElection (DISPOSE) approach to render the person by using an appropriate mesh for a given pose. Unlike pose space animation methods that require manual or offline time consuming pose set creation, our distortion score based scheme can choose the mesh to be transmitted and update the pose set accordingly. DISPOSE works with partial meshes and does not require dense registration enabling real time pose space creation. With DISPOSE incorporated into 3DTI, the latency for rendering the mesh on the receiving side is limited by only the transmission delay of the skeletal data (which is only around 250 bytes). Our evaluations show the effectiveness of DISPOSE for generating good quality online animation faster than real time.

3D Tele-Immersion (3DTI) environments are emerging as a new medium for human interactions and collaborations in the areas of education, sports training, physical medicine and rehabilitation. By adding a tactile element to a visually centered 3DTI environment, such applications can be made even more engaging. But it also opens up a few challenges in terms of fusing the visual and tactile data streams in a synchronous way. In this paper we describe a 3DTI Tele-Rehabilitation system with Microsoft Kinect cameras and hap tic devices. We describe some of the challenges we face in providing as well as quantifying a good quality of experience (QoE) in this system. We propose a set of solutions that: (i) improve the user's QoE (by using multi-modal prediction for handling latencies, better synchronization that accounts for the global state of the system, etc.), (ii) quantify the QoE (by designing a controlled virtual environment and by defining appropriate user QoE metrics for immersive tele-rehabilitation). The experimental results show a marked improvement in the performance of the system, consequently improving the user-experience. This is also verified by the results of the user performance study.

Cardiopulmonary resuscitation (CPR) plays a primary role in first-aid treatment. Instead of the traditional instructor-led training course, we propose a virtual reality system which provides an immersive 3D environment for CPR training with visual and haptic feedback. To simulate a real world CPR experience, our immersive trainer system enables a trainee to perform CPR compressions to a virtual human, inside the virtual world. During the training procedure, the trainee can not only watch his/her 3D image performing CPR, but also feel the force feedback from the chest compressions in real-time. To further enhance the visual fidelity, a haptic-enabled deformable model is applied to show the visual change of chest during compression.

3D collaborative Tele-Immersive environments allow reconstruction of real world 3D scenes in the virtual world across multiple physical locations. This kind of reconstruction results in a lot of 3D data being transmitted over the internet in real time. The current systems allow for transmission at low frame rates due to the large volume of data and network bandwidth restrictions. In this paper we propose a prediction based approach that generates future frames by animating the live model based on few skeleton points. By doing so the magnitude of data transmitted is reduced to few hundred bytes. The prediction errors are corrected when an entire frame is received. This approach allows minimal amounts (few bytes) of data to be transmitted per frame, thus allowing for high frame rates and still maintain an acceptable visual quality of reconstruction at the receiver side.

In recent years, many local descriptor based approaches have been proposed for human activity recognition, which perform well on challenging datasets. However, most of these approaches are computationally intensive, extract irrelevant background features and fail to capture global temporal information. We propose to overcome these issues by introducing a compact and robust motion space that can be used to extract both spatial and temporal aspects of activities using local descriptors. We present Speed Adapted Motion History Image Space (SAMHIS) that employs a variant of Motion History Image for representing motion. This space alleviates both self-occlusion as well as the speed-related issues associated with different kinds of motion. We go on to show using a standard bag of visual words model that extracting appearance based local descriptors from this space is very effective for recognizing activity. Our approach yields promising results on the KTH and Weizmann dataset.

We present an immersive gaming demonstration using the minimum amount of wearable sensors. The game demonstrated is two-player tennis. We combine a virtual environment with real 3D representations of physical objects like the players and the tennis racquet (if available). The main objective of the game is to provide as real an experience of tennis as possible, while also being as less intrusive as possible. The game is played across a network, and this opens the possibility of two remote players playing a game together on a single virtual tennis pitch. The Microsoft Kinect sensors are used to obtain a 3D point cloud and a skeletal map representation of the player. This 3D point cloud is mapped on to the virtual tennis pitch. We also use a wireless wearable Attitude and Heading Reference System (AHRS) mote, which is strapped onto the wrist of the players. This mote gives us precise information about the movement (swing, rotation etc.) of the playing arm. This information along with the skeletal map is used to implement the physics of the game. Using this game we demonstrate our solutions for simultaneous data acquisition, 3D point-cloud mapping in a virtual space, use of the Kinect and AHRS sensors to calibrate real and virtual objects and for interaction of virtual objects with a 3D point cloud.