We introduce a method for learning adversarial perturbations targeted to individual images or videos. The learned perturbations are found to be sparse while at the same time containing a high level of feature detail. Thus, the extracted perturbations allow a form of object or action recognition and provide insights into what features the studied deep neural network models consider important when reaching their classification decisions. From an adversarial point of view, the sparse perturbations successfully confused the models into misclassifying, although the perturbed samples still belonged to the same original class by visual examination. This is discussed in terms of a prospective data augmentation scheme. The sparse yet high-quality perturbations may also be leveraged for image or video compression.
Inter-frame Coding plays significant role for video Compression and Computer Vision. Computer vision systems have been incorporated in many real life applications (e.g. surveillance systems, medical imaging, robot navigation and identity verification systems). Object tracking is a key computer vision topic, which aims at detecting the position of a moving object from a video sequence. The application of Inter-frame Coding for low frame rate video, as well as for low resolution video. Various methods based on Top-down approach just like kernel based or mean shift technique are used to track the object for video, So, Inter-frame Coding algorithms are widely adopted by video coding standards, mainly due to their simplicity and good distortion performance for object tracking.
Video processing systems such as HEVC requiring low energy consumption needed for the multimedia market has lead to extensive development in fast algorithms for the efficient approximation of 2-D DCT transforms. The DCT is employed in a multitude of compression standards due to its remarkable energy compaction properties. Multiplier-free approximate DCT transforms have been proposed that offer superior compression performance at very low circuit complexity. Such approximations can be realized in digital VLSI hardware using additions and subtractions only, leading to significant reductions in chip area and power consumption compared to conventional DCTs and integer transforms. In this paper, we introduce a novel 8-point DCT approximation that requires only 14 addition operations and no multiplications. The proposed transform possesses low computational complexity and is compared to state-of-the-art DCT approximations in terms of both algorithm complexity and peak signal-to-noise ratio. The proposed DCT approximation is a candidate for reconfigurable video standards such as HEVC. The proposed transform and several other DCT approximations are mapped to systolic-array digital architectures and physically realized as digital prototype circuits using FPGA technology and mapped to 45 nm CMOS technology.
Sparse decomposition has been extensively used for different applications including signal compression and denoising and document analysis. In this paper, sparse decomposition is used for image segmentation. The proposed algorithm separates the background and foreground using a sparse-smooth decomposition technique such that the smooth and sparse components correspond to the background and foreground respectively. This algorithm is tested on several test images from HEVC test sequences and is shown to have superior performance over other methods, such as the hierarchical k-means clustering in DjVu. This segmentation algorithm can also be used for text extraction, video compression and medical image segmentation.
In many mobile visual analysis applications, compressed video is transmitted over a communication network and analyzed by a server. Typical processing steps performed at the server include keypoint detection, descriptor calculation, and feature matching. Video compression has been shown to have an adverse effect on feature-matching performance. The negative impact of compression can be reduced by using the keypoints extracted from the uncompressed video to calculate descriptors from the compressed video. Based on this observation, we propose to provide these keypoints to the server as side information and to extract only the descriptors from the compressed video. First, we introduce four different frame types for keypoint encoding to address different types of changes in video content. These frame types represent a new scene, the same scene, a slowly changing scene, or a rapidly moving scene and are determined by comparing features between successive video frames. Then, we propose Intra, Skip and Inter modes of encoding the keypoints for different frame types. For example, keypoints for new scenes are encoded using the Intra mode, and keypoints for unchanged scenes are skipped. As a result, the bitrate of the side information related to keypoint encoding is significantly reduced. Finally, we present pairwise matching and image retrieval experiments conducted to evaluate the performance of the proposed approach using the Stanford mobile augmented reality dataset and 720p format videos. The results show that the proposed approach offers significantly improved feature matching and image retrieval performance at a given bitrate.
It is a high cost problem for panoramic image stitching via image matching algorithm and not practical for real-time performance. In this paper, we take full advantage ofHarris corner invariant characterization method light intensity parallel meaning, translation and rotation, and made a realtime panoramic image stitching algorithm. According to the basic characteristics and performance FPGA classical algorithm, several modules such as the feature point extraction, and matching description is to optimize the feature-based logic. Real-time optimization system to achieve high precision match. The new algorithm process the image from pixel domain and obtained from CCD camera Xilinx Spartan-6 hardware platform. After the image stitching algorithm, will eventually form a portable interface to output high-definition content on the display. The results showed that, the proposed algorithm has higher precision with good real-time performance and robustness.
Robust face detection in the wild is one of the ultimate components to support various facial related problems, i.e. unconstrained face recognition, facial periocular recognition, facial landmarking and pose estimation, facial expression recognition, 3D facial model construction, etc. Although the face detection problem has been intensely studied for decades with various commercial applications, it still meets problems in some real-world scenarios due to numerous challenges, e.g. heavy facial occlusions, extremely low resolutions, strong illumination, exceptionally pose variations, image or video compression artifacts, etc. In this paper, we present a face detection approach named Contextual Multi-Scale Region-based Convolution Neural Network (CMS-RCNN) to robustly solve the problems mentioned above. Similar to the region-based CNNs, our proposed network consists of the region proposal component and the region-of-interest (RoI) detection component. However, far apart of that network, there are two main contributions in our proposed network that play a significant role to achieve the state-of-the-art performance in face detection. Firstly, the multi-scale information is grouped both in region proposal and RoI detection to deal with tiny face regions. Secondly, our proposed network allows explicit body contextual reasoning in the network inspired from the intuition of human vision system. The proposed approach is benchmarked on two recent challenging face detection databases, i.e. the WIDER FACE Dataset which contains high degree of variability, as well as the Face Detection Dataset and Benchmark (FDDB). The experimental results show that our proposed approach trained on WIDER FACE Dataset outperforms strong baselines on WIDER FACE Dataset by a large margin, and consistently achieves competitive results on FDDB against the recent state-of-the-art face detection methods.
A multiplierless pruned approximate 8-point discrete cosine transform (DCT) requiring only 10 additions is introduced. The proposed algorithm was assessed in image and video compression, showing competitive performance with state-of-the-art methods. Digital implementation in 45 nm CMOS technology up to place-and-route level indicates clock speed of 288 MHz at a 1.1 V supply. The 8x8 block rate is 36 MHz.The DCT approximation was embedded into HEVC reference software; resulting video frames, at up to 327 Hz for 8-bit RGB HEVC, presented negligible image degradation.
It has long been considered a significant problem to improve the visual quality of lossy image and video compression. Recent advances in computing power together with the availability of large training data sets has increased interest in the application of deep learning cnns to address image recognition and image processing tasks. Here, we present a powerful cnn tailored to the specific task of semantic image understanding to achieve higher visual quality in lossy compression. A modest increase in complexity is incorporated to the encoder which allows a standard, off-the-shelf jpeg decoder to be used. While jpeg encoding may be optimized for generic images, the process is ultimately unaware of the specific content of the image to be compressed. Our technique makes jpeg content-aware by designing and training a model to identify multiple semantic regions in a given image. Unlike object detection techniques, our model does not require labeling of object positions and is able to identify objects in a single pass. We present a new cnn architecture directed specifically to image compression, which generates a map that highlights semantically-salient regions so that they can be encoded at higher quality as compared to background regions. By adding a complete set of features for every class, and then taking a threshold over the sum of all feature activations, we generate a map that highlights semantically-salient regions so that they can be encoded at a better quality compared to background regions. Experiments are presented on the Kodak PhotoCD dataset and the MIT Saliency Benchmark dataset, in which our algorithm achieves higher visual quality for the same compressed size.
The usage of linear transformations has great relevance for data decorrelation applications, like image and video compression. In that sense, the discrete Tchebichef transform (DTT) possesses useful coding and decorrelation properties. The DTT transform kernel does not depend on the input data and fast algorithms can be developed to real time applications. However, the DTT fast algorithm presented in literature possess high computational complexity. In this work, we introduce a new low-complexity approximation for the DTT. The fast algorithm of the proposed transform is multiplication-free and requires a reduced number of additions and bit-shifting operations. Image and video compression simulations in popular standards shows good performance of the proposed transform. Regarding hardware resource consumption for FPGA shows 43.1% reduction of configurable logic blocks and ASIC place and route realization shows 57.7% reduction in the area-time figure when compared with the 2-D version of the exact DTT.
Traditional image and video compression algorithms rely on hand-crafted encoder/decoder pairs (codecs) that lack adaptability and are agnostic to the data being compressed. Here we describe the concept of generative compression, the compression of data using generative models, and suggest that it is a direction worth pursuing to produce more accurate and visually pleasing reconstructions at much deeper compression levels for both image and video data. We also demonstrate that generative compression is orders-of-magnitude more resilient to bit error rates (e.g. from noisy wireless channels) than traditional variable-length coding schemes.
Rapid and low power computation of optical flow (OF) is potentially useful in robotics. The dynamic vision sensor (DVS) event camera produces quick and sparse output, and has high dynamic range, but conventional OF algorithms are frame-based and cannot be directly used with event-based cameras. Previous DVS OF methods do not work well with dense textured input and are designed for implementation in logic circuits. This paper proposes a new block-matching based DVS OF algorithm which is inspired by motion estimation methods used for MPEG video compression. The algorithm was implemented both in software and on FPGA. For each event, it computes the motion direction as one of 9 directions. The speed of the motion is set by the sample interval. Results show that the Average Angular Error can be improved by 30\% compared with previous methods. The OF can be calculated on FPGA with 50\,MHz clock in 0.2\,us per event (11 clock cycles), 20 times faster than a Java software implementation running on a desktop PC. Sample data is shown that the method works on scenes dominated by edges, sparse features, and dense texture.
Cell tracking enables data extraction from time-lapse "cell movies" and promotes modeling biological processes at the single-cell level. We introduce a new fully automated computational strategy to track accurately cells across frames in time-lapse movies. Our method is based on a dynamic neighborhoods formation and matching approach, inspired by motion estimation algorithms for video compression. Moreover, it exploits "divide and conquer" opportunities to solve effectively the challenging cells tracking problem in overcrowded bacterial colonies. Using cell movies generated by different labs we demonstrate that the accuracy of the proposed method remains very high (exceeds 97%) even when analyzing large overcrowded microbial colonies.
Saliency detection has drawn a lot of attention of researchers in various fields over the past several years. Saliency is the perceptual quality that makes an object, person to draw the attention of humans at the very sight. Salient object detection in an image has been used centrally in many computational photography and computer vision applications like video compression, object recognition and classification, object segmentation, adaptive content delivery, motion detection, content aware resizing, camouflage images and change blindness images to name a few. We propose a method to detect saliency in the objects using multimodal dictionary learning which has been recently used in classification and image fusion. The multimodal dictionary that we are learning is task driven which gives improved performance over its counterpart (the one which is not task specific).
Sampling is a fundamental topic in graph signal processing, having found applications in estimation, clustering, and video compression. In contrast to traditional signal processing, the irregularity of the signal domain makes selecting a sampling set non-trivial and hard to analyze. Indeed, though conditions for graph signal interpolation from noiseless samples exist, they do not lead to a unique sampling set. The presence of noise makes choosing among these sampling sets a hard combinatorial problem. Although greedy sampling schemes are commonly used in practice, they have no performance guarantee. This work takes a twofold approach to address this issue. First, universal performance bounds are derived for the Bayesian estimation of graph signals from noisy samples. In contrast to currently available bounds, they are not restricted to specific sampling schemes and hold for any sampling sets. Second, this paper provides near-optimal guarantees for greedy sampling by introducing the concept of approximate submodularity and updating the classical greedy bound. It then provides explicit bounds on the approximate supermodularity of the interpolation mean-square error showing that it can be optimized with worst-case guarantees using greedy search even though it is not supermodular. Simulations illustrate the derived bound for different graph models and show an application of graph signal sampling to reduce the complexity of kernel principal component analysis.
We present DeepCache, a principled cache design for deep learning inference in continuous mobile vision. DeepCache benefits model execution efficiency by exploiting temporal locality in input video streams. It addresses a key challenge raised by mobile vision: the cache must operate under video scene variation, while trading off among cacheability, overhead, and loss in model accuracy. At the input of a model, DeepCache discovers video temporal locality by exploiting the video's internal structure, for which it borrows proven heuristics from video compression; into the model, DeepCache propagates regions of reusable results by exploiting the model's internal structure. Notably, DeepCache eschews applying video heuristics to model internals which are not pixels but high-dimensional, difficult-to-interpret data. Our implementation of DeepCache works with unmodified deep learning models, requires zero developer's manual effort, and is therefore immediately deployable on off-the-shelf mobile devices. Our experiments show that DeepCache saves inference execution time by 18% on average and up to 47%. DeepCache reduces system energy consumption by 20% on average.
We study domain-specific video streaming. Specifically, we target a streaming setting where the videos to be streamed from a server to a client are all in the same domain and they have to be compressed to a small size for low-latency transmission. Several popular video streaming services, such as the video game streaming services of GeForce Now and Twitch, fall in this category. While conventional video compression standards such as H.264 are commonly used for this task, we hypothesize that one can leverage the property that the videos are all in the same domain to achieve better video quality. Based on this hypothesis, we propose a novel video compression pipeline. Specifically, we first apply H.264 to compress domain-specific videos. We then train a novel binary autoencoder to encode the leftover domain-specific residual information frame-by-frame into binary representations. These binary representations are then compressed and sent to the client together with the H.264 stream. In our experiments, we show that our pipeline yields consistent gains over standard H.264 compression across several benchmark datasets while using the same channel bandwidth.
There has been a growing interest in using different approaches to improve the coding efficiency of modern video codec in recent years as demand for web-based video consumption increases. In this paper, we propose a model-based approach that uses texture analysis/synthesis to reconstruct blocks in texture regions of a video to achieve potential coding gains using the AV1 codec developed by the Alliance for Open Media (AOM). The proposed method uses convolutional neural networks to extract texture regions in a frame, which are then reconstructed using a global motion model. Our preliminary results show an increase in coding efficiency while maintaining satisfactory visual quality.
Image and video compression has traditionally been tailored to human vision. However, modern applications such as visual analytics and surveillance rely on computers seeing and analyzing the images before (or instead of) humans. For these applications, it is important to adjust compression to computer vision. In this paper we present a bit allocation and rate control strategy that is tailored to object detection. Using the initial convolutional layers of a state-of-the-art object detector, we create an importance map that can guide bit allocation to areas that are important for object detection. The proposed method enables bit rate savings of 7% or more compared to default HEVC, at the equivalent object detection rate.
Training robust deep video representations has proven to be much more challenging than learning deep image representations. This is in part due to the enormous size of raw video streams and the high temporal redundancy; the true and interesting signal is often drowned in too much irrelevant data. Motivated by that the superfluous information can be reduced by up to two orders of magnitude by video compression (using H.264, HEVC, etc.), we propose to train a deep network directly on the compressed video. This representation has a higher information density, and we found the training to be easier. In addition, the signals in a compressed video provide free, albeit noisy, motion information. We propose novel techniques to use them effectively. Our approach is about 4.6 times faster than Res3D and 2.7 times faster than ResNet-152. On the task of action recognition, our approach outperforms all the other methods on the UCF-101, HMDB-51, and Charades dataset.