Search

Pursuing more complete and coherent scene understanding towards realistic vision applications drives edge detection from category-agnostic to category-aware semantic level. However, finer delineation of instance-level boundaries still remains unexcavated. In this work, we address a new finer-grained task, termed panoptic edge detection (PED), which aims at predicting semantic-level boundaries for stuff categories and instance-level boundaries for instance categories, in order to provide more comprehensive and unified scene understanding from the perspective of edges.We then propose a versatile framework, Panoptic Edge Network (PEN), which aggregates different tasks of object detection, semantic and instance edge detection into a single holistic network with multiple branches. Based on the same feature representation, the semantic edge branch produces semantic-level boundaries for all categories and the object detection branch generates instance proposals. Conditioned on the prior information from these two branches, the instance edge branch aims at instantiating edge predictions for instance categories. Besides, we also devise a Panoptic Dual F-measure (F2) metric for the new PED task to uniformly measure edge prediction quality for both stuff and instances. By joint end-to-end training, the proposed PEN framework outperforms all competitive baselines on Cityscapes and ADE20K datasets.

Generative adversarial nets (GANs) and variational auto-encoders have significantly improved our distribution modeling capabilities, showing promise for dataset augmentation, image-to-image translation and feature learning. However, to model high-dimensional distributions, sequential training and stacked architectures are common, increasing the number of tunable hyper-parameters as well as the training time. Nonetheless, the sample complexity of the distance metrics remains one of the factors affecting GAN training. We first show that the recently proposed sliced Wasserstein distance has compelling sample complexity properties when compared to the Wasserstein distance. To further improve the sliced Wasserstein distance we then analyze its `projection complexity' and develop the max-sliced Wasserstein distance which enjoys compelling sample complexity while reducing projection complexity, albeit necessitating a max estimation. We finally illustrate that the proposed distance trains GANs on high-dimensional images up to a resolution of 256x256 easily., Comment: Accepted to CVPR 2019

The design of multi-functional BMGs is limited by the lack of a quantitative understanding of the variables that control the glass-forming ability (GFA) of alloys. Both geometric frustration (e.g. differences in atomic radii) and energetic frustration (e.g. differences in the cohesive energies of the atomic species) contribute to the GFA. We perform molecular dynamics simulations of binary Lennard-Jones mixtures with only energetic frustration. We show that there is little correlation between the heat of mixing and critical cooling rate $R_c$, below which the system crystallizes, except that $\Delta H_{\rm mix} < 0$. By removing the effects of geometric frustration, we show strong correlations between $R_c$ and the variables $\epsilon_- = (\epsilon_{BB}-\epsilon_{AA})/(\epsilon_{AA}+\epsilon_{BB})$ and ${\overline \epsilon}_{AB} = 2\epsilon_{AB}/(\epsilon_{AA}+\epsilon_{BB})$, where $\epsilon_{AA}$ and $\epsilon_{BB}$ are the cohesive energies of atoms $A$ and $B$ and $\epsilon_{AB}$ is the pair interaction between $A$ and $B$ atoms. We identify a particular $f_B$-dependent combination of $\epsilon_-$ and ${\overline \epsilon}_{AB}$ that collapses the data for $R_c$ over nearly $4$ orders of magnitude in cooling rate.

Features from multiple scales can greatly benefit the semantic edge detection task if they are well fused. However, the prevalent semantic edge detection methods apply a fixed weight fusion strategy where images with different semantics are forced to share the same weights, resulting in universal fusion weights for all images and locations regardless of their different semantics or local context. In this work, we propose a novel dynamic feature fusion strategy that assigns different fusion weights for different input images and locations adaptively. This is achieved by a proposed weight learner to infer proper fusion weights over multi-level features for each location of the feature map, conditioned on the specific input. In this way, the heterogeneity in contributions made by different locations of feature maps and input images can be better considered and thus help produce more accurate and sharper edge predictions. We show that our model with the novel dynamic feature fusion is superior to fixed weight fusion and also the na\"ive location-invariant weight fusion methods, via comprehensive experiments on benchmarks Cityscapes and SBD. In particular, our method outperforms all existing well established methods and achieves new state-of-the-art.