Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

/pdf/deep-residual-learning-for-image-recognition-b75jg61qrq.pdf

Deep Residual Learning for Image Recognition

In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

Very Deep Convolutional Networks for Large-Scale Image Recognition

Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers---8x deeper than VGG nets but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. 
The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

Deep learning is a form of machine learning that enables computers to learn from experience and understand the world in terms of a hierarchy of concepts. Because the computer gathers knowledge from experience, there is no need for a human computer operator to formally specify all the knowledge that the computer needs. The hierarchy of concepts allows the computer to learn complicated concepts by building them out of simpler ones; a graph of these hierarchies would be many layers deep. This book introduces a broad range of topics in deep learning. The text offers mathematical and conceptual background, covering relevant concepts in linear algebra, probability theory and information theory, numerical computation, and machine learning. It describes deep learning techniques used by practitioners in industry, including deep feedforward networks, regularization, optimization algorithms, convolutional networks, sequence modeling, and practical methodology; and it surveys such applications as natural language processing, speech recognition, computer vision, online recommendation systems, bioinformatics, and videogames. Finally, the book offers research perspectives, covering such theoretical topics as linear factor models, autoencoders, representation learning, structured probabilistic models, Monte Carlo methods, the partition function, approximate inference, and deep generative models. Deep Learning can be used by undergraduate or graduate students planning careers in either industry or research, and by software engineers who want to begin using deep learning in their products or platforms. A website offers supplementary material for both readers and instructors.

Deep Learning

Research in EMG-based control of prostheses has mainly utilized adult subjects who have fully developed neuromuscular control. Little is known about children’s ability to generate consistent EMG signals necessary to control artificial limbs with multiple degrees of freedom. As a first step to address this gap, experiments were designed to validate and benchmark two experimental protocols that quantify the ability to coordinate forearm muscle contractions in typically developing children. Non-disabled, healthy adults and children participated in our experiments that aimed to measure an individual’s ability to use myoelectric control interfaces. In the first experiment, participants performed 8 repetitions of 16 different hand/wrist movements. Using offline classification analysis based on Support Vector Machine, we quantified their ability to consistently produce distinguishable muscle contraction patterns. We demonstrated that children had a smaller number of highly independent movements (can be classified with >90% accuracy) than adults did. The second experiment measured participants’ ability to control the position of a cursor on a 1-DoF virtual slide using proportional EMG control with three different visuomotor gain levels. We found that children had higher failure rates and slower average target acquisitions than adults did, primarily due to longer correction times that did not improve over repetitive practice. We also found that the performance in both experiments was age-dependent in children. The results of this study provide novel insights into the technical and empirical basis to better understand neuromuscular development in children with upper-limb loss.

https://ieeexplore.ieee.org/ielx7/7333/4359219/09755884.pdf

Age-Dependent Upper Limb Myoelectric Control Capability in Typically Developing Children

Contrary to the vast literature in modeling, perceiving, and understanding agent-object (e.g., human-object, hand-object, robot-object) interaction in computer vision and robotics, very few past works have studied the task of object-object interaction, which also plays an important role in robotic manipulation and planning tasks. There is a rich space of object-object interaction scenarios in our daily life, such as placing an object on a messy tabletop, fitting an object inside a drawer, pushing an object using a tool, etc. In this paper, we propose a unified affordance learning framework to learn object-object interaction for various tasks. By constructing four object-object interaction task environments using physical simulation (SAPIEN) and thousands of ShapeNet models with rich geometric diversity, we are able to conduct large-scale object-object affordance learning without the need for human annotations or demonstrations. At the core of technical contribution, we propose an object-kernel point convolution network to reason about detailed interaction between two objects. Experiments on large-scale synthetic data and real-world data prove the effectiveness of the proposed approach. Please refer to the project webpage for code, data, video, and more materials: this https URL

/pdf/o2o-afford-annotation-free-large-scale-object-object-3uqpo4l2k2.pdf

O2O-Afford: Annotation-Free Large-Scale Object-Object Affordance Learning.

In deepwater-drilling engineering, it is necessary to disconnect the bottom equipment of the lower marine-riser package from the blowout preventer when encountering multi-hazard environmental factors. In order to reduce the impact of recoil on the drilling platform after the sudden disconnection of the riser, in this paper, an optimal guaranteed cost H∞ recoil control problem is considered for the drilling riser. First, a three-element mass-damper-spring deepwater-drilling riser model subject to fluid discharge and heave motion of offshore platform is given. Then, an optimal guaranteed cost H∞ controller (OGCHC) is designed to suppress the recoil response of the drilling riser, and the sufficient conditions for the asymptotic stability of the closed-loop system are derived. Third, it is found through simulation results that the designed OGCHC can reduce the recoil response effectively. In order to further analyze the advantages of the OGCHC, the performance indices of the riser without active-recoil control and with optimal control (OC) and OGCHC are compared. It is shown that the average response amplitudes of three mass blocks of the riser are almost the same, while the control cost by the OGCHC is less than that by the OC. Further, under the designed recoil control, no riser compression occurs, thereby ensuring the safety of the riser system.

/pdf/recoil-control-of-deepwater-drilling-riser-with-optimal-2c90kruj.pdf

Recoil Control of Deepwater-Drilling Riser with Optimal Guaranteed Cost H∞ Control

We introduce GNeRF, a framework to marry Generative Adversarial Networks (GAN) with Neural Radiance Field (NeRF) reconstruction for the complex scenarios with unknown and even randomly initialized camera poses. Recent NeRF-based advances have gained popularity for remarkable realistic novel view synthesis. However, most of them heavily rely on accurate camera poses estimation, while few recent methods can only optimize the unknown camera poses in roughly forward-facing scenes with relatively short camera trajectories and require rough camera poses initialization. Differently, our GNeRF only utilizes randomly initialized poses for complex outside-in scenarios. We propose a novel two-phases end-to-end framework. The first phase takes the use of GANs into the new realm for optimizing coarse camera poses and radiance fields jointly, while the second phase refines them with additional photometric loss. We overcome local minima using a hybrid and iterative optimization scheme. Extensive experiments on a variety of synthetic and natural scenes demonstrate the effectiveness of GNeRF. More impressively, our approach outperforms the baselines favorably in those scenes with repeated patterns or even low textures that are regarded as extremely challenging before.

GNeRF: GAN-based Neural Radiance Field without Posed Camera

Many applications of unpaired image-to-image translation require the input contents to be preserved semantically during translations. Unaware of the inherently unmatched semantics distributions between source and target domains, existing distribution matching methods (i.e., GAN-based) can give undesired solutions. In particular, although producing visually reasonable outputs, the learned models usually flip the semantics of the inputs. To tackle this without using extra supervision, we propose to enforce the translated outputs to be semantically invariant w.r.t. small perceptual variations of the inputs, a property we call "semantic robustness". By optimizing a robustness loss w.r.t. multi-scale feature space perturbations of the inputs, our method effectively reduces semantics flipping and produces translations that outperform existing methods both quantitatively and qualitatively.

Hao Su

Papers

Age-Dependent Upper Limb Myoelectric Control Capability in Typically Developing Children

O2O-Afford: Annotation-Free Large-Scale Object-Object Affordance Learning.

Recoil Control of Deepwater-Drilling Riser with Optimal Guaranteed Cost H∞ Control

GNeRF: GAN-based Neural Radiance Field without Posed Camera

Semantically Robust Unpaired Image Translation for Data with Unmatched Semantics Statistics.