Model compression techniques are recently gaining explosive attention for obtaining efficient AI models for various real-time applications. Channel pruning is one important compression strategy and is widely used in slimming various DNNs. Previous ga
te-based or importance-based pruning methods aim to remove channels whose importance is smallest. However, it remains unclear what criteria the channel importance should be measured on, leading to various channel selection heuristics. Some other sampling-based pruning methods deploy sampling strategies to train sub-nets, which often causes the training instability and the compressed models degraded performance. In view of the research gaps, we present a new module named Gates with Differentiable Polarization (GDP), inspired by principled optimization ideas. GDP can be plugged before convolutional layers without bells and whistles, to control the on-and-off of each channel or whole layer block. During the training process, the polarization effect will drive a subset of gates to smoothly decrease to exact zero, while other gates gradually stay away from zero by a large margin. When training terminates, those zero-gated channels can be painlessly removed, while other non-zero gates can be absorbed into the succeeding convolution kernel, causing completely no interruption to training nor damage to the trained model. Experiments conducted over CIFAR-10 and ImageNet datasets show that the proposed GDP algorithm achieves the state-of-the-art performance on various benchmark DNNs at a broad range of pruning ratios. We also apply GDP to DeepLabV3Plus-ResNet50 on the challenging Pascal VOC segmentation task, whose test performance sees no drop (even slightly improved) with over 60% FLOPs saving.
Analyzing and understanding hand information from multimedia materials like images or videos is important for many real world applications and remains active in research community. There are various works focusing on recovering hand information from
single image, however, they usually solve a single task, for example, hand mask segmentation, 2D/3D hand pose estimation, or hand mesh reconstruction and perform not well in challenging scenarios. To further improve the performance of these tasks, we propose a novel Hand Image Understanding (HIU) framework to extract comprehensive information of the hand object from a single RGB image, by jointly considering the relationships between these tasks. To achieve this goal, a cascaded multi-task learning (MTL) backbone is designed to estimate the 2D heat maps, to learn the segmentation mask, and to generate the intermediate 3D information encoding, followed by a coarse-to-fine learning paradigm and a self-supervised learning strategy. Qualitative experiments demonstrate that our approach is capable of recovering reasonable mesh representations even in challenging situations. Quantitatively, our method significantly outperforms the state-of-the-art approaches on various widely-used datasets, in terms of diverse evaluation metrics.
We propose ResRep, a novel method for lossless channel pruning (a.k.a. filter pruning), which slims down a CNN by reducing the width (number of output channels) of convolutional layers. Inspired by the neurobiology research about the independence of
remembering and forgetting, we propose to re-parameterize a CNN into the remembering parts and forgetting parts, where the former learn to maintain the performance and the latter learn to prune. Via training with regular SGD on the former but a novel update rule with penalty gradients on the latter, we realize structured sparsity. Then we equivalently merge the remembering and forgetting parts into the original architecture with narrower layers. In this sense, ResRep can be viewed as a successful application of Structural Re-parameterization. Such a methodology distinguishes ResRep from the traditional learning-based pruning paradigm that applies a penalty on parameters to produce sparsity, which may suppress the parameters essential for the remembering. ResRep slims down a standard ResNet-50 with 76.15% accuracy on ImageNet to a narrower one with only 45% FLOPs and no accuracy drop, which is the first to achieve lossless pruning with such a high compression ratio. The code and models are at https://github.com/DingXiaoH/ResRep.
Color compatibility is important for evaluating the compatibility of a fashion outfit, yet it was neglected in previous studies. We bring this important problem to researchers attention and present a compatibility learning framework as solution to va
rious fashion tasks. The framework consists of a novel way to model outfit compatibility and an innovative learning scheme. Specifically, we model the outfits as graphs and propose a novel graph construction to better utilize the power of graph neural networks. Then we utilize both ground-truth labels and pseudo labels to train the compatibility model in a weakly-supervised manner.Extensive experimental results verify the importance of color compatibility alone with the effectiveness of our framework. With color information alone, our models performance is already comparable to previous methods that use deep image features. Our full model combining the aforementioned contributions set the new state-of-the-art in fashion compatibility prediction.
We present a palette-based framework for color composition for visual applications. Color composition is a critical aspect of visual applications in art, design, and visualization. The color wheel is often used to explain pleasing color combinations
in geometric terms, and, in digital design, to provide a user interface to visualize and manipulate colors. We abstract relationships between palette colors as a compact set of axes describing harmonic templates over perceptually uniform color wheels. Our framework provides a basis for a variety of color-aware image operations, such as color harmonization and color transfer, and can be applied to videos. To enable our approach, we introduce an extremely scalable and efficient yet simple palette-based image decomposition algorithm. Our approach is based on the geometry of images in RGBXY-space. This new geometric approach is orders of magnitude more efficient than previous work and requires no numerical optimization. We demonstrate a real-time layer decomposition tool. After preprocessing, our algorithm can decompose 6 MP images into layers in 20 milliseconds. We also conducted three large-scale, wide-ranging perceptual studies on the perception of harmonic colors and harmonization algorithms.
The colorful appearance of a physical painting is determined by the distribution of paint pigments across the canvas, which we model as a per-pixel mixture of a small number of pigments with multispectral absorption and scattering coefficients. We pr
esent an algorithm to efficiently recover this structure from an RGB image, yielding a plausible set of pigments and a low RGB reconstruction error. We show that under certain circumstances we are able to recover pigments that are close to ground truth, while in all cases our results are always plausible. Using our decomposition, we repose standard digital image editing operations as operations in pigment space rather than RGB, with interestingly novel results. We demonstrate tonal adjustments, selection masking, cut-copy-paste, recoloring, palette summarization, and edge enhancement.
In digital painting software, layers organize paintings. However, layers are not explicitly represented, transmitted, or published with the final digital painting. We propose a technique to decompose a digital painting into layers. In our decompositi
on, each layer represents a coat of paint of a single paint color applied with varying opacity throughout the image. Our decomposition is based on the paintings RGB-space geometry. In RGB-space, a geometric structure is revealed due to the linear nature of the standard Porter-Duff over pixel compositing operation. The vertices of the convex hull of pixels in RGB-space suggest paint colors. Users choose the degree of simplification to perform on the convex hull, as well as a layer order for the colors. We solve a constrained optimization problem to find maximally translucent, spatially coherent opacity for each layer, such that the composition of the layers reproduces the original image. We demonstrate the utility of the resulting decompositions for re-editing.
