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

There is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization. We show that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Using the same network trained on transmitted light microscopy images (phase contrast and DIC) we won the ISBI cell tracking challenge 2015 in these categories by a large margin. Moreover, the network is fast. Segmentation of a 512x512 image takes less than a second on a recent GPU. The full implementation (based on Caffe) and the trained networks are available at http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net .

/pdf/u-net-convolutional-networks-for-biomedical-image-2xoyjzggxv.pdf

U-Net: Convolutional Networks for Biomedical Image Segmentation

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.

This paper introduces an extremely efficient CNN architecture named DFANet for semantic segmentation under resource constraints. Our proposed network starts from a single lightweight backbone and aggregates discriminative features through sub-network and sub-stage cascade respectively. Based on the multi-scale feature propagation, DFANet substantially reduces the number of parameters, but still obtains sufficient receptive field and enhances the model learning ability, which strikes a balance between the speed and segmentation performance. Experiments on Cityscapes and CamVid datasets demonstrate the superior performance of DFANet with 8$\times$ less FLOPs and 2$\times$ faster than the existing state-of-the-art real-time semantic segmentation methods while providing comparable accuracy. Specifically, it achieves 70.3\% Mean IOU on the Cityscapes test dataset with only 1.7 GFLOPs and a speed of 160 FPS on one NVIDIA Titan X card, and 71.3\% Mean IOU with 3.4 GFLOPs while inferring on a higher resolution image.

DFANet: Deep Feature Aggregation for Real-Time Semantic Segmentation

Previous video stabilization methods often employ homographies to model transitions between consecutive frames, or require robust long feature tracks. However, the homography model is invalid for scenes with significant depth variations, and feature point tracking is fragile in videos with textureless objects, severe occlusion or camera rotation. To address these challenging cases, we propose to solve video stabilization with an additional depth sensor such as the Kinect camera. Though the depth image is noisy, incomplete and low resolution, it facilitates both camera motion estimation and frame warping, which make the video stabilization a much well posed problem. The experiments demonstrate the effectiveness of our algorithm.

/pdf/video-stabilization-with-a-depth-camera-5brmx4ji1r.pdf

Video stabilization with a depth camera

In this paper, we propose a novel query design for the transformer-based object detection. 
In previous transformer-based detectors, the object queries are a set of learned embeddings.
However, each learned embedding does not have an explicit physical meaning and we cannot explain where it will focus on.
It is difficult to optimize as the prediction slot of each object query does not have a specific mode. 
In other words, each object query will not focus on a specific region.
To solve these problems, in our query design, object queries are based on anchor points, which are widely used in CNN-based detectors. 
So each object query focuses on the objects near the anchor point. 
Moreover, our query design can predict multiple objects at one position to solve the difficulty: ``one region, multiple objects''.
In addition, we design an attention variant, which can reduce the memory cost while achieving similar or better performance than the standard attention in DETR.
Thanks to the query design and the attention variant, the proposed detector that we called Anchor DETR, can achieve better performance and run faster than the DETR with 10x fewer training epochs.
For example, it achieves 44.2 AP with 19 FPS on the MSCOCO dataset when using the ResNet50-DC5 feature for training 50 epochs.
Extensive experiments on the MSCOCO benchmark prove the effectiveness of the proposed methods.
Code is available at https://github.com/megvii-research/AnchorDETR.

/pdf/anchor-detr-query-design-for-transformer-based-detector-2ouhyvqz.pdf

Anchor DETR: Query Design for Transformer-Based Detector

Many applications that process a point cloud data benefit from a reliable normal estimation step. Given a point cloud presumably sampled from an unknown surface, the problem is to estimate the normals of the surface at the data points. Two approaches, one based on numerical optimizations and another based on Voronoi diagrams are known for the problem. Variations of numerical approaches work well even when point clouds are contaminated with noise. Recently a variation of the Voronoi based method is proposed for noisy point clouds. The centrality of the normal estimation step in point cloud processing begs a thorough study of the two approaches so that one knows which approach is appropriate for what circumstances. This paper presents such results.

Normal estimation for point clouds: a comparison study for a Voronoi based method

Object detectors are usually equipped with networks designed for image classification as backbones, e.g., ResNet. Although it is publicly known that there is a gap between the task of image classification and object detection, designing a suitable detector backbone is still manually exhaustive. In this paper, we propose DetNAS to automatically search neural architectures for the backbones of object detectors. In DetNAS, the search space is formulated into a supernet and the search method relies on evolution algorithm (EA). In experiments, we show the effectiveness of DetNAS on various detectors, the one-stage detector, RetinaNet, and the twostage detector, FPN. For each case, we search in both training from scratch scheme and ImageNet pre-training scheme. There is a consistent superiority compared to the architectures searched on ImageNet classification. Our main result architecture achieves better performance than ResNet-101 on COCO with the FPN detector. In addition, we illustrate the architectures searched by DetNAS and find some meaningful patterns.

Jian Sun

Papers

DFANet: Deep Feature Aggregation for Real-Time Semantic Segmentation

Video stabilization with a depth camera

Anchor DETR: Query Design for Transformer-Based Detector

Normal estimation for point clouds: a comparison study for a Voronoi based method

DetNAS: Neural Architecture Search on Object Detection.