In this paper we propose an innovative framework able to extract the perimeter of the rooftop of buildings from remote sensing images. The core features are derived from an existing algorithm which performs an automatic rooftop extraction in nadir aerial imagery. We significantly improve its performance by adding new characteristics in order to speed up the whole framework and as a result the system can achieve better performance in terms of both complexity reduction and capability of handling heterogeneous data sets.

Standard video coding systems currently employed in UAV (Unmanned Aerial Vehicle) and aerial drone applications do not rely on some peculiarities in terms of scene 3D model and correlation among successive frames. In particular, the observed scene is static, i.e. the camera movement is dominant, and it can often be well approximated with a plane. Moreover, camera position and orientation can be obtained from the navigation system. Therefore, correspondent points on two video frames are linked by a simple homography. This paper presents novel results obtained by a low-complexity sensor aided H.264 encoder, recently developed at CIRA and yet tested on simulated data. The proposed encoder employs a new motion estimation scheme which make use of the global motion information provided by the onboard navigation system. The homography is used in order to initialize the block matching algorithm allowing a more robust motion estimation and a smaller search window, and hence reducing the complexity. The tests are made coding real aerial imagery, captured to be used for 3D scene reconstruction. The images are acquired by an high resolution camera mounted on a small drone, flying at low altitude.

In this paper we propose an innovative video-based architecture aimed to monitor elderly people. It is based on cheap devices and open-source libraries and preliminary tests demonstrate that it manages to achieve a significant performance. The overall architecture of the system and its implementation are shortly discussed from the point of view of the composing functional blocks, also analyzing the effects of loopbacks on the effectiveness of the algorithm.

This paper presents a new low-complexity H.264 encoder for Unmanned Aerial Vehicles (UAV) applications. Standard video coding systems currently employed in UAV applications do not rely on some peculiarities in terms of scene 3D model and correlation among successive frames. In particular, the observed scene is static, i.e. the camera movement is dominant, and it can often be well approximated with a plane. Moreover, camera position and orientation can be obtained from the navigation system. Therefore, correspondent points on two video frames are linked by a simple homography. The encoder employs a new motion estimation scheme which make use of the global motion information provided by the onboard navigation system. The homography is used in order to initialize the block matching algorithm allowing a more robust motion estimation and a smaller search window, and hence reducing the complexity. Experimental results show that the proposed scheme ouperforms standard H.264 in terms of PSNR and throughput. The results are relevant in low frame rate video coding, which is a typical scenario in UAV behind line-of-sight (BLOS) missions. Experiments open new drections in developing new sensor aided video coding standards.

In this paper, we devise a procedure that mimics the behavior of a progressive video stream starting from a non progressive one such as H.264/AVC encoded video. This allows one to unequally protect the video data in an efficient way, according to their importance and the network state. The reported results demonstrate the superior performance of the proposed approach in comparison to state-of-the-art methods for resilient transmission of H.264/AVC data. Moreover, the flexibility in terms of redundancy insertion and achieved quality levels, allows one to span different applications, possibly including P2P video streaming.

We propose a novel standard-compliant multiple description coding (MDC) method that exploits the H.264/AVC redundant slice tool, performing rate-distortion optimization at the slice level. The strategy to allocate redundancy to each slice jointly takes into account its contribution to distortion, its position in the GOP, the effect of decoder error concealment, and the transmission conditions. This makes the algorithm more accurate with respect to previous frame-based solutions, and experimental results show that it compares favorably with other state-of-the-art standard-compliant MDC techniques.

In this letter, we present a performance comparison between multiple description coding (MDC) and unequal loss protection (ULP) for progressive image transmission over lossy packet networks. Two optimization criteria are considered, i.e., a multi-quality criterion, whenNdistinct quality levels are guaranteed at the decoder side, and the optimization of the expected quality at the receiver. We resort to both a semi-analytical approach and simulation results. To enable numerical comparisons, we address a specific MDC algorithm suitable for progressive imaging, and a state-of-the-art ULP algorithm based on Reed Solomon codes. The results, although cannot be generalized to any MDC and ULP methods, are useful to put into evidence some general features that can drive the selection of the most proper technique for the application at hand. In fact, they allow to put into evidence the main advantages and drawbacks of either technique.

Multiple description coding (MDC) makes use of redundant representations of multimedia data to achieve resiliency. Descriptions should be generated so that the quality obtained when decoding a subset of them only depends on their number and not on the particular received subset. In this paper, we propose a method based on the principle of encoding the source at several rates, and properly blending the data encoded at different rates to generate the descriptions. The aim is to achieve efficient redundancy exploitation, and easy adaptation to different network scenarios by means of fine tuning of the encoder parameters. We apply this principle to both JPEG 2000 images and H.264/AVC video data. We consider as the reference scenario the distribution of contents on application-layer overlays with multiple-tree topology. The experimental results reveal that our method favorably compares with state-of-art MDC techniques.

In this paper we address a novel scheme to protect video sequences according to slice importance based on slice reordering, ULP and error-concealment techniques. The approach does not require the modification to the video decoder although an application-layer channel coding is required. Simulation results show that the proposed algorithm outperforms state-of-the-art approaches, reducing the gap with the upper-bound error-free performance curve. Moreover, the complexity of the additional stage required to pilot the protection allocation stage is negligible with respect to traditional ULP schemes.

In this letter, we propose a novel unequal loss protection scheme, which allocates FEC codes to video slices according to their impact on the GOP distortion. This is evaluated taking the concealment procedure and the drift effect into account. Simulation results show that the proposed algorithm outperforms state-of-the-art approaches, reducing the gap with the error-free performance curve. Moreover, the complexity of the additional stage required to pilot the protection allocation stage is negligible with respect to traditional ULP schemes.

Multiple description coding (MDC) is a good way to combat packet losses in error-prone networks subject to packet erasures. However, redundancy tuning is often difficult, and this makes the generation of descriptions with good redundancy-rate-distortion performance a hard job. Moreover, the complexity of generating more than two descriptions represents a strong limitation to MDC. In this letter, we propose a method that exploits the data rate-distortion characteristics to generate multiple descriptions for JPEG 2000 with tunable redundancy levels. The identification of the best number of descriptions as a function of the network conditions is also addressed

The efficient transmission of video content over wireless communication networks is a challenging goal, especially when considering multiple mobile users equipped with energy-constrained handheld devices and sharing the same channel resources. In the present work, considering the scenario of the real-time delivery of multiple MJPEG2000 video streams over a QoS-enabled 802.11e wireless LAN, we investigate the performance of multiple description coding and compare it with different video transmission techniques. We analyze if and when data redundancy of the information source can improve the end-to-end perceived quality in this realistic multi-user scenario

Despite multiple description coding (MDC) paradigm is effective to combat packet losses on networks, it is not included in any multimedia co-decoding standard. This can be justified by the complexity of the redundancy tuning task of many of the proposed methods. Moreover, the complexity of generating more than two descriptions represents a strong limitation to MDC. In this paper, we generalize the MDC scheme proposed for JPEG 2000 in T. Tillo and G. Olmo (2004) to more than two descriptions. Using this flexible MDC scheme, we show the importance of properly tuning both the redundancy and the number of descriptions to match the network condition. Finally, we derive the conditions in order to determine when it is better to use single description coding or MDC

Sensor networks are usually dense networks where the network diversity can be exploited in order to overcome failures. In this paper, we study the use of multiple description techniques in the context of sensor networks where the cause of failures is due to the usual practical constraint of having finite buffers in the sensors, instead of the more traditional case of link failures considered in previous research. Although from a theoretical point of view we observe that the use of more descriptions provides usually better performance, we show experimentally that this is not the case in practice, when real constraints are introduced, such as finite buffers and the presence of header information, necessary for any real application. Our main result is that the optimal number of descriptions, in terms of average distortion, decreases as the fraction of header information increases for a given buffer size

When transmitting multimedia information over non-prioritized networks subject to packet losses, multiple description coding with an arbitrary number of descriptions is an effective choice in order to minimize the end-to-end distortion. Such descriptions should be generated so that the quality obtained decoding a subset of them depends only on their number and not on the particular received subset. In this paper, we propose an encoding procedure to generate an arbitrary number of balanced descriptions using a multi-rate allocation scheme which exploits the R-D characteristic of the data