Search

Both performance and efficiency are important to semantic segmentation. State-of-the-art semantic segmentation algorithms are mostly based on dilated Fully Convolutional Networks (dilatedFCN), which adopt dilated convolutions in the backbone networks to extract high-resolution feature maps for achieving high-performance segmentation performance. However, due to many convolution operations are conducted on the high-resolution feature maps, such dilatedFCN-based methods result in large computational complexity and memory consumption. To balance the performance and efficiency, there also exist encoder-decoder structures that gradually recover the spatial information by combining multi-level feature maps from the encoder. However, the performances of existing encoder-decoder methods are far from comparable with the dilatedFCN-based methods. In this paper, we propose the EfficientFCN, whose backbone is a common ImageNet pre-trained network without any dilated convolution. A holistically-guided decoder is introduced to obtain the high-resolution semantic-rich feature maps via the multi-scale features from the encoder. The decoding task is converted to novel codebook generation and codeword assembly task, which takes advantages of the high-level and low-level features from the encoder. Such a framework achieves comparable or even better performance than state-of-the-art methods with only 1/3 of the computational cost. Extensive experiments on PASCAL Context, PASCAL VOC, ADE20K validate the effectiveness of the proposed EfficientFCN., Comment: Accepted in ECCV 2020

The acceleration of super-heavy ions (SHIs) from plasmas driven by ultrashort (tens of femtoseconds) laser pulses is a challenging topic waiting for breakthrough. The detecting and controlling of the ionization process, and the adoption of the optimal acceleration scheme are crucial for the generation of highly energetic SHIs. Here, we report the experimental results on the generation of deeply ionized super-heavy ions (Au) with unprecedented energy of 1.2 GeV utilizing ultrashort laser pulses (22 fs) at the intensity of 10^22 W/cm2. A novel self-calibrated diagnostic method was developed to acquire the absolute energy spectra and charge state distributions of Au ions abundant at the charge state of 51+ and reaching up to 61+. The measured charge state distributions supported by 2D particle-in-cell simulations serves as an additional tool to inspect the ionization dynamics associated with SHI acceleration, revealing that the laser intensity is the crucial parameter for the acceleration of Au ions over the pulse duration. The use of double-layer targets results in a prolongation of the acceleration time without sacrificing the strength of acceleration field, which is highly favorable for the generation of high-energy super heavy ions.

Long-range contextual information is essential for achieving high-performance semantic segmentation. Previous feature re-weighting methods demonstrate that using global context for re-weighting feature channels can effectively improve the accuracy of semantic segmentation. However, the globally-sharing feature re-weighting vector might not be optimal for regions of different classes in the input image. In this paper, we propose a Context-adaptive Convolution Network (CaC-Net) to predict a spatially-varying feature weighting vector for each spatial location of the semantic feature maps. In CaC-Net, a set of context-adaptive convolution kernels are predicted from the global contextual information in a parameter-efficient manner. When used for convolution with the semantic feature maps, the predicted convolutional kernels can generate the spatially-varying feature weighting factors capturing both global and local contextual information. Comprehensive experimental results show that our CaC-Net achieves superior segmentation performance on three public datasets, PASCAL Context, PASCAL VOC 2012 and ADE20K., Comment: Accepted in ECCV 2020

Artistic style transfer can be thought as a process to generate different versions of abstraction of the original image. However, most of the artistic style transfer operators are not optimized for human faces thus mainly suffers from two undesirable features when applying them to selfies. First, the edges of human faces may unpleasantly deviate from the ones in the original image. Second, the skin color is far from faithful to the original one which is usually problematic in producing quality selfies. In this paper, we take a different approach and formulate this abstraction process as a gradient domain learning problem. We aim to learn a type of abstraction which not only achieves the specified artistic style but also circumvents the two aforementioned drawbacks thus highly applicable to selfie photography. We also show that our method can be directly generalized to videos with high inter-frame consistency. Our method is also robust to non-selfie images, and the generalization to various kinds of real-life scenes is discussed. We will make our code publicly available.

We report a positioning method for transparent targets with an accuracy of \SI{2}{\mu m} for a compact laser proton accelerator. The positioning system consists of two light-emitting diodes (LED), a long working distance objective and two charge coupled devices (CCD) for illumination, imaging and detection, respectively. We developed a defocusing method making transparent targets visible as phase objects and applied it to our system. Precise positioning of transparent targets can be realized by means of minimizing the image contrast of the phase objects. Fast positioning based on the relationship between the radius of spherical aberration ring and defocusing distance is also realized. Laser proton acceleration experiments have been performed to demonstrate the reliability of this positioning system., Comment: 5 pages, 4 figures

We observed that recent state-of-the-art results on single image human pose estimation were achieved by multi-stage Convolution Neural Networks (CNN). Notwithstanding the superior performance on static images, the application of these models on videos is not only computationally intensive, it also suffers from performance degeneration and flicking. Such suboptimal results are mainly attributed to the inability of imposing sequential geometric consistency, handling severe image quality degradation (e.g. motion blur and occlusion) as well as the inability of capturing the temporal correlation among video frames. In this paper, we proposed a novel recurrent network to tackle these problems. We showed that if we were to impose the weight sharing scheme to the multi-stage CNN, it could be re-written as a Recurrent Neural Network (RNN). This property decouples the relationship among multiple network stages and results in significantly faster speed in invoking the network for videos. It also enables the adoption of Long Short-Term Memory (LSTM) units between video frames. We found such memory augmented RNN is very effective in imposing geometric consistency among frames. It also well handles input quality degradation in videos while successfully stabilizes the sequential outputs. The experiments showed that our approach significantly outperformed current state-of-the-art methods on two large-scale video pose estimation benchmarks. We also explored the memory cells inside the LSTM and provided insights on why such mechanism would benefit the prediction for video-based pose estimations., Comment: Poster in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018