Recent character image animation methods based on diffusion models, such as Animate Anyone, have made significant progress in generating consistent and generalizable character animations. However, these approaches fail to produce reasonable associations between characters and their environments. To address this limitation, we introduce Animate Anyone 2, aiming to animate characters with environment affordance. Beyond extracting motion signals from source video, we additionally capture environmental representations as conditional inputs. The environment is formulated as the region with the exclusion of characters and our model generates characters to populate these regions while maintaining coherence with the environmental context. We propose a shape-agnostic mask strategy that more effectively characterizes the relationship between character and environment. Furthermore, to enhance the fidelity of object interactions, we leverage an object guider to extract features of interacting objects and employ spatial blending for feature injection. We also introduce a pose modulation strategy that enables the model to handle more diverse motion patterns. Experimental results demonstrate the superior performance of the proposed method.

Existing actor-critic algorithms, which are popular for continuous control reinforcement learning (RL) tasks, suffer from poor sample efficiency due to lack of principled exploration mechanism within them. Motivated by the success of Thompson sampling for efficient exploration in RL, we propose a novel model-free RL algorithm, \emph{Langevin Soft Actor Critic} (LSAC), which prioritizes enhancing critic learning through uncertainty estimation over policy optimization. LSAC employs three key innovations: approximate Thompson sampling through distributional Langevin Monte Carlo (LMC) based $Q$ updates, parallel tempering for exploring multiple modes of the posterior of the $Q$ function, and diffusion synthesized state-action samples regularized with $Q$ action gradients. Our extensive experiments demonstrate that LSAC outperforms or matches the performance of mainstream model-free RL algorithms for continuous control tasks. Notably, LSAC marks the first successful application of an LMC based Thompson sampling in continuous control tasks with continuous action spaces.

We present a novel method for reconstructing clothed humans from a sparse set of, e.g., 1–6 RGB images. Despite impressive results from recent works employing deep implicit representation, we revisit the volumetric approach and demonstrate that better performance can be achieved with proper system design. The volumetric representation offers significant advantages in leveraging 3D spatial context through 3D convolutions, and the notorious quantization error is largely negligible with a reasonably large yet affordable volume resolution, e.g., 512. To handle memory and computation costs, we propose a sophisticated coarse-to-fine strategy with voxel culling and subspace sparse convolution. Our method starts with a discretized visual hull to compute a coarse shape and then focuses on a narrow band nearby the coarse shape for refinement. Once the shape is reconstructed, we adopt an image-based rendering approach, which computes the colors of surface points by blending input images with learned weights. Extensive experimental results show that our method significantly reduces the mean point-to-surface (P2S) precision of state-of-the-art methods by more than 50% to achieve approximately 2mm accuracy with a 512 volume resolution. Additionally, images rendered from our textured model achieve a higher peak signal-to-noise ratio (PSNR) compared to state-of-the-art methods.

Abstract This paper focuses on high efficiency secret key generation mechanism of physical‐layer communication over fading channels in ubiquitous wireless networks. The secret key rate via traditional physical‐layer approach could be limited when the wireless propagation channels connecting two sensors change slowly. To generate a high‐rate secret key and improve the communication efficiency over quasi‐static block fading channels, a novel multi‐randomness device‐to‐device secret key generation strategy and a cooperative communication mechanism aided by relay nodes are proposed. In the proposed schemes, the legitimate members to send random signals rotationally in every coherent time are set; thus, two legitimate ubiquitous wireless network members, Alice and Bob, can obtain the potential correlated information by exploiting the randomness and the reciprocity of the wireless propagation channels. Considering the reciprocity of wireless channels is variable while the forward channel gain and backward channel gain are correlated in coherent time, a modified secret key generation scheme is proposed via layered coding with theoretical secret key rates derived. The simulation results show that the proposed scheme outperforms traditional approaches with favourable application prospects in ubiquitous wireless communications networks and internet of things.

Mining a set of meaningful and distinctive topics automatically from massive text corpora has broad applications. Existing topic models, however, typically work in a purely unsupervised way, which often generate topics that do not fit users’ particular needs and yield suboptimal performance on downstream tasks. We propose a new task, discriminative topic mining, which leverages a set of user-provided category names to mine discriminative topics from text corpora. This new task not only helps a user understand clearly and distinctively the topics he/she is most interested in, but also benefits directly keyword-driven classification tasks. We develop CatE, a novel category-name guided text embedding method for discriminative topic mining, which effectively leverages minimal user guidance to learn a discriminative embedding space and discover category representative terms in an iterative manner. We conduct a comprehensive set of experiments to show that CatE mines high-quality set of topics guided by category names only, and benefits a variety of downstream applications including weakly-supervised classification and lexical entailment direction identification.

Emergency case reasoning is essential to emergency management. In this paper, we propose a novel emergency case reasoning method based on fuzzy cognitive map (FCM), to model the inherent causal relationships in emergency cases. Specifically, we first obtain emergency domain elements and mine their association rules, by leveraging natural language processing technology and FT-Growth dada mining algorithm. We then design an effective algorithm to learn causal knowledge links from the gathered association rules. Finally, we construct an FCM regarding emergency events and show the reasoning process. Experiments on the gas explosion demonstrate that the proposed method can successfully model the internal causal relationships of emergency elements, and the development of the emergency event can be reflected by the reasoning process of the proposed method according to its varying variables of state. The proposed method can effectively inference and predict the tendency of emergency cases based on the reasoning process, which can further provide valuable decision supports to emergency responders.

Unsupervised text embedding has shown great power in a wide range of NLP tasks. While text embeddings are typically learned in the Euclidean space, directional similarity is often more effective in tasks such as word similarity and document clustering, which creates a gap between the training stage and usage stage of text embedding. To close this gap, we propose a spherical generative model based on which unsupervised word and paragraph embeddings are jointly learned. To learn text embeddings in the spherical space, we develop an efficient optimization algorithm with convergence guarantee based on Riemannian optimization. Our model enjoys high efficiency and achieves state-of-the-art performances on various text embedding tasks including word similarity and document clustering.

Motivated by applications in social networks, a new type of dominating set named Positive Influence Dominating Set (PIDS) has been studied in the literature. In this paper, we investigate cooperative cost games arising from PIDS problem on social network graphs. We propose two new game models, Rigid PIDS Game and Relaxed PIDS Game, and focus on their cores. First, a relationship between the cores of both games is obtained. Next, we also prove that the core of the relaxed PIDS game is nonempty if and only if there is no integrality gap for the relaxation linear programming of the PIDS problem on graph G.

A lot of self-stabilizing algorithms for computing dominating sets problem have been proposed in the literature due to many real-life applications. Most of the proposed algorithms either work for dominating sets with a uniform weight or find approximation solutions to weighted dominating sets. However, for non-uniform weighted dominating sets (WDS) problem, there is no self-stabilizing algorithm for the WDS. Furthermore, how to find the minimal weighted dominating set is a challenge. In this paper, we propose a self-stabilizing algorithm for the minimal weighted dominating set (MWDS) under a central daemon model when operating in any general network. We further prove that the worst case convergence time of the algorithm from any arbitrary initial state is O(n2) steps where n is the number of nodes in the network.