From the Publisher: 
The accessible presentation of this book gives both a general view of the entire computer vision enterprise and also offers sufficient detail to be able to build useful applications. Users learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods. A CD-ROM with every copy of the text contains source code for programming practice, color images, and illustrative movies. Comprehensive and up-to-date, this book includes essential topics that either reflect practical significance or are of theoretical importance. Topics are discussed in substantial and increasing depth. Application surveys describe numerous important application areas such as image based rendering and digital libraries. Many important algorithms broken down and illustrated in pseudo code. Appropriate for use by engineers as a comprehensive reference to the computer vision enterprise.

Computer vision : a modern approach = 计算机视觉 : 一种现代的方法

Learning Deconvolution Network for Semantic Segmentation

Deep learning on image denoising: An overview.

A survey on deep learning and its applications

We introduce a novel network, called as CO-attention Siamese Network (COSNet), to address the unsupervised video object segmentation task from a holistic view. We emphasize the importance of inherent correlation among video frames and incorporate a global co-attention mechanism to improve further the state-of-the-art deep learning based solutions that primarily focus on learning discriminative foreground representations over appearance and motion in short-term temporal segments. The co-attention layers in our network provide efficient and competent stages for capturing global correlations and scene context by jointly computing and appending co-attention responses into a joint feature space. We train COSNet with pairs of video frames, which naturally augments training data and allows increased learning capacity. During the segmentation stage, the co-attention model encodes useful information by processing multiple reference frames together, which is leveraged to infer the frequently reappearing and salient foreground objects better. We propose a unified and end-to-end trainable framework where different co-attention variants can be derived for mining the rich context within videos. Our extensive experiments over three large benchmarks manifest that COSNet outperforms the current alternatives by a large margin. We will publicly release our implementation and models.

/pdf/see-more-know-more-unsupervised-video-object-segmentation-3x53tihu4o.pdf

See More, Know More: Unsupervised Video Object Segmentation With Co-Attention Siamese Networks

In this paper, we propose a novel MoNet model to deeply exploit motion cues for boosting video object segmentation performance from two aspects, i.e., frame representation learning and segmentation refinement. Concretely, MoNet exploits computed motion cue (i.e., optical flow) to reinforce the representation of the target frame by aligning and integrating representations from its neighbors. The new representation provides valuable temporal contexts for segmentation and improves robustness to various common contaminating factors, e.g., motion blur, appearance variation and deformation of video objects. Moreover, MoNet exploits motion inconsistency and transforms such motion cue into foreground/background prior to eliminate distraction from confusing instances and noisy regions. By introducing a distance transform layer, MoNet can effectively separate motion-inconstant instances/regions and thoroughly refine segmentation results. Integrating the proposed two motion exploitation components with a standard segmentation network, MoNet provides new state-of-the-art performance on three competitive benchmark datasets.

/pdf/monet-deep-motion-exploitation-for-video-object-segmentation-1ut5alz5b2.pdf

MoNet: Deep Motion Exploitation for Video Object Segmentation

Image demosaicing and denoising are import steps of image signal processing. Sequential executions of demosaicing and denoising have essential drawbacks that they degrade the results of each other. Joint demosaicing and denoising overcomes the difficulties by solving the two problems in one model. This paper introduces a unified object function with hidden priors and a variant of ADMM to recover a full-resolution color image with a noisy Bayer input. Experimental results demonstrate that our method performs better than state-of-the-art methods in both PSNR comparison and human vision. In addition, our method is much more robust to variations of noise level.

Joint demosaicing and denoising of noisy bayer images with ADMM

Unconventional emissions from excitons and trions in monolayer WS2 are studied by photoexcitation. When excited by a 532 nm laser beam, the carrier species in the monolayer WS2 are affected by the excess electrons escaping from photoionization of donor impurity, the concentration of which varies with different locations of the sample. Simply by increasing the excitation power at room temperature, the excess electrons and, thus, the intensity ratio of excited trions and excitons can be continuously tuned over a large range from 0.1 to 7.7. Furthermore, this intensity ratio can also be manipulated by varying temperature. However, in this way, the resonance energy of the excitons and trions shows redshifts with increasing temperature due to electron–phonon coupling. The binding energy of the trion is determined to be ∼26 meV and independent of temperature, indicating strong Coulomb interaction of carriers in such 2D materials.

Large range modification of exciton species in monolayer WS 2 .

Monolayer (ML) transition metal dichalcogenides (TMDCs) are thought to be highly promising materials for the optoelectronic and nanophotonic applications. However, the low absorption cross section and photoluminescence (PL) quantum yield in such atomically thin layers restrict their applications. Considering that the energy transfer in a heterostructure can modulate TMDCs’ optical properties, a type I heterostructure geometry comprising ML TMDCs and lead halide perovskite quantum dots (QDs) has the potential to overcome these shortcomings. In this work, spin-coating the CsPbBr3 QDs on ML WS2 results in ~12.7 times enhancement in the PL intensity of ML WS2 at 295K. This giant enhancement is attributed to the energy transfer process from CsPbBr3 QDs to WS2 with a ~40% energy transfer efficiency and ~2 × 108 s−1 energy transfer rate. Besides, we observed that the internal quantum efficiency of ML WS2 is increased from 6.35% to 29.01%. The result demonstrates the feasibility of using perovskite QDs and ML TMDCs to form a type I heterostructure and improve the performance of the TMDC-based optoelectronic devices.

Giant photoluminescence enhancement in monolayer WS_2 by energy transfer from CsPbBr_3 quantum dots

Conventional deep neural networks based video object segmentation (VOS) methods are dominated by heavily fine-tuning a segmentation model on the first frame of a given video, which is time-consuming and inefficient. In this paper, we propose a novel method which rapidly adapts a base segmentation model to new video sequences with only a couple of model-update iterations, without sacrificing performance. Such attractive efficiency benefits from the meta-learning paradigm which leads to a meta-segmentation model and a novel continuous learning approach which enables online adaptation of the segmentation model. Concretely, we train a meta-learner on multiple VOS tasks such that the meta model can capture their common knowledge and gains the ability to fast adapt the segmentation model to new video sequences. Furthermore, to deal with unique challenges of VOS tasks from temporal variations in the video, e.g., object motion and appearance changes, we propose a principled online adaptation approach that continuously adapts the segmentation model across video frames by exploiting temporal context effectively, providing robustness to annoying temporal variations. Integrating the meta-learner with the online adaptation approach, the proposed VOS model achieves competitive performance against the state-of-the-arts and moreover provides faster per-frame processing speed.

Yu Liu

Papers

MoNet: Deep Motion Exploitation for Video Object Segmentation

Joint demosaicing and denoising of noisy bayer images with ADMM

Large range modification of exciton species in monolayer WS 2 .

Giant photoluminescence enhancement in monolayer WS_2 by energy transfer from CsPbBr_3 quantum dots

Online Meta Adaptation for Fast Video Object Segmentation