Floors are Flat: Leveraging Semantics for Real-Time Surface Normal Prediction

We propose 4 insights that help to significantly improve the performance of deep learning models that predict surface normals and semantic labels from a single RGB image. These insights are: (1) denoise the "ground truth" surface normals in the training set to ensure consistency with the semantic labels; (2) concurrently train on a mix of real and synthetic data, instead of pretraining on synthetic and finetuning on real; (3) jointly predict normals and semantics using a shared model, but only backpropagate errors on pixels that have valid training labels; (4) slim down the model and use grayscale instead of color inputs. Despite the simplicity of these steps, we demonstrate consistently improved results on several datasets, using a model that runs at 12 fps on a standard mobile phone.

Tracking Emerges by Colorizing Videos

We use large amounts of unlabeled video to learn models for visual tracking without manual human supervision. We leverage the natural temporal coherency of color to create a model that learns to colorize gray-scale videos by copying colors from a reference frame. Quantitative and qualitative experiments suggest that this task causes the model to automatically learn to track visual regions. Although the model is trained without any ground-truth labels, our method learns to track well enough to outperform the latest methods based on optical flow. Moreover, our results suggest that failures to track are correlated with failures to colorize, indicating that advancing video colorization may further improve self-supervised visual tracking.

Instance Embedding Transfer to Unsupervised Video Object Segmentation

We propose a method for unsupervised video object segmentation by transferring the knowledge encapsulated in image-based instance embedding networks. The instance embedding network produces an embedding vector for each pixel that enables identifying all pixels belonging to the same object. Though trained on static images, the instance embeddings are stable over consecutive video frames, which allows us to link objects together over time. Thus, we adapt the instance networks trained on static images to video object segmentation and incorporate the embeddings with objectness and optical flow features, without model retraining or online fine-tuning. The proposed method outperforms state-of-the-art unsupervised segmentation methods in the DAVIS dataset and the FBMS dataset.

The Devil is in the Decoder

Many machine vision applications require predictions for every pixel of the input image (for example semantic segmentation, boundary detection). Models for such problems usually consist of encoders which decreases spatial resolution while learning a high-dimensional representation, followed by decoders who recover the original input resolution and result in low-dimensional predictions. While encoders have been studied rigorously, relatively few studies address the decoder side. Therefore this paper presents an extensive comparison of a variety of decoders for a variety of pixel-wise prediction tasks. Our contributions are: (1) Decoders matter: we observe significant variance in results between different types of decoders on various problems. (2) We introduce a novel decoder: bilinear additive upsampling. (3) We introduce new residual-like connections for decoders. (4) We identify two decoder types which give a consistently high performance.

Speed/accuracy trade-offs for modern convolutional object detectors

The goal of this paper is to serve as a guide for selecting a detection architecture that achieves the right speed/memory/accuracy balance for a given application and platform. To this end, we investigate various ways to trade accuracy for speed and memory usage in modern convolutional object detection systems. A number of successful systems have been proposed in recent years, but apples-to-apples comparisons are difficult due to different base feature extractors (e.g., VGG, Residual Networks), different default image resolutions, as well as different hardware and software platforms. We present a unified implementation of the Faster R-CNN [Ren et al., 2015], R-FCN [Dai et al., 2016] and SSD [Liu et al., 2015] systems, which we view as "meta-architectures" and trace out the speed/accuracy trade-off curve created by using alternative feature extractors and varying other critical parameters such as image size within each of these meta-architectures. On one extreme end of this spectrum where speed and memory are critical, we present a detector that achieves real time speeds and can be deployed on a mobile device. On the opposite end in which accuracy is critical, we present a detector that achieves state-of-the-art performance measured on the COCO detection task.

Semantic Instance Segmentation via Deep Metric Learning

We propose a new method for semantic instance segmentation, by first computing how likely two pixels are to belong to the same object, and then by grouping similar pixels together. Our similarity metric is based on a deep, fully convolutional embedding model. Our grouping method is based on selecting all points that are sufficiently similar to a set of "seed points", chosen from a deep, fully convolutional scoring model. We show competitive results on the Pascal VOC instance segmentation benchmark.

VideoSET: Video Summary Evaluation through Text

In this paper we present VideoSET, a method for Video Summary Evaluation through Text that can evaluate how well a video summary is able to retain the semantic information contained in its original video. We observe that semantics is most easily expressed in words, and develop a text-based approach for the evaluation. Given a video summary, a text representation of the video summary is first generated, and an NLP-based metric is then used to measure its semantic distance to ground-truth text summaries written by humans. We show that our technique has higher agreement with human judgment than pixel-based distance metrics. We also release text annotations and ground-truth text summaries for a number of publicly available video datasets, for use by the computer vision community.

An introduction to synchronous self-learning Pareto strategy

In last decades optimization and control of complex systems that possessed various conflicted objectives simultaneously attracted an incremental interest of scientists. This is because of the vast applications of these systems in various fields of real life engineering phenomena that are generally multi modal, non convex and multi criterion. Hence, many researchers utilized versatile intelligent models such as Pareto based techniques, game theory (cooperative and non cooperative games), neuro evolutionary systems, fuzzy logic and advanced neural networks for handling these types of problems. In this paper a novel method called Synchronous Self Learning Pareto Strategy Algorithm (SSLPSA) is presented which utilizes Evolutionary Computing (EC), Swarm Intelligence (SI) techniques and adaptive Classical Self Organizing Map (CSOM) simultaneously incorporating with a data shuffling behavior. Evolutionary Algorithms (EA) which attempt to simulate the phenomenon of natural evolution are powerful numerical optimization algorithms that reach an approximate global maximum of a complex multi variable function over a wide search space and swarm base technique can improved the intensity and the robustness in EA. CSOM is a neural network capable of learning and can improve the quality of obtained optimal Pareto front. To prove the efficient performance of proposed algorithm, authors utilized some well known benchmark test functions. Obtained results indicate that the cited method is best suit in the case of vector optimization.

A natural-inspired optimization machine based on the annual migration of salmons in nature

Bio inspiration is a branch of artificial simulation science that shows pervasive contributions to variety of engineering fields such as automate pattern recognition, systematic fault detection and applied optimization. In this paper, a new metaheuristic optimizing algorithm that is the simulation of The Great Salmon Run(TGSR) is developed. The obtained results imply on the acceptable performance of implemented method in optimization of complex non convex, multi dimensional and multi-modal problems. To prove the superiority of TGSR in both robustness and quality, it is also compared with most of the well known proposed optimizing techniques such as Simulated Annealing (SA), Parallel Migrating Genetic Algorithm (PMGA), Differential Evolutionary Algorithm (DEA), Particle Swarm Optimization (PSO), Bee Algorithm (BA), Artificial Bee Colony (ABC), Firefly Algorithm (FA) and Cuckoo Search (CS). The obtained results confirm the acceptable performance of the proposed method in both robustness and quality for different bench-mark optimizing problems and also prove the authors claim.