We present Animus3D , a text-driven 3D animation framework that generates motion field given a static 3D asset and text prompt. Previous methods mostly leverage the vanilla Score Distillation Sampling (SDS) objective to distill motion from pretrained text-to-video diffusion, leading to animations with minimal movement or noticeable jitter. To address this, our approach introduces a novel SDS alternative, Motion Score Distillation (MSD). Specifically, we introduce a LoRA-enhanced video diffusion model that defines a static source distribution rather than pure noise as in SDS, while another inversion-based noise estimation technique ensures appearance preservation when guiding motion. To further improve motion fidelity, we incorporate explicit temporal and spatial regularization terms that mitigate geometric distortions across time and space. Additionally, we propose a motion refinement module to upscale the temporal resolution and enhance fine-grained details, overcoming the fixed-resolution constraints of the underlying video model. Extensive experiments demonstrate that Animus3D successfully animates static 3D assets from diverse text prompts, generating significantly more substantial and detailed motion than state-of-the-art baselines while maintaining high visual integrity. Code will be released upon acceptance.

In recent years, sparse voxel-based methods have become the state-of-the-arts for 3D semantic segmentation of indoor scenes, thanks to the powerful 3D CNNs. Nevertheless, being oblivious to the underlying geometry, voxel-based methods suffer from ambiguous features on spatially close objects and struggle with handling complex and irregular geometries due to the lack of geodesic information. In view of this, we present Voxel-Mesh Network (VMNet), a novel 3D deep architecture that operates on the voxel and mesh representations leveraging both the euclidean and geodesic information. Intuitively, the euclidean information extracted from voxels can offer contextual cues representing interactions between nearby objects, while the geodesic information extracted from meshes can help separate objects that are spatially close but have disconnected surfaces. To incorporate such information from the two domains, we design an intra-domain attentive module for effective feature aggregation and an inter-domain attentive module for adaptive feature fusion. Experimental results validate the effectiveness of VMNet: specifically, on the challenging ScanNet dataset for large-scale segmentation of indoor scenes, it outperforms the state-of-the-art SparseConvNet and MinkowskiNet (74.6% versus 72.5% and 73.6% in mIoU) with a simpler network structure (17M versus 30M and 38M parameters).

Recent advancements in human fashion video generation have transformed the field, producing various promising effects. Existing methods mainly focus on pose control but lack the ability to achieve sketch-based control, largely due to the absence of appearance-consistent and shape-varying knowledge in existing datasets. Moreover, the necessity of sequential structure inputs to control video generation hinders real-world applications. To address these limitations, we introduce Sketch2HumanVideo, an approach that, for the first time, achieves sketch-controllable human video generation with three conditions: temporally sparse sketches, a spatially sparse pose sequence, and a reference appearance image. Our key contribution is a sparse sketch encoder, which takes the first two conditions as input, enabling precise and multi-view control of shape motion. To provide the above knowledge, we leverage the expertise of two pretrained models to synthesize a dataset comprising shape-varying yet appearance-consistent examples for model training. Furthermore, we introduce an enlarging-and-resampling scheme to enhance high-frequency details of local regions in resource-constrained scenarios, thereby promoting the generation of realistic videos. Through qualitative and quantitative experiments, our method showcases superior performance to state-of-the-art approaches and flexible control.

Geometry- and appearance-controlled full-body human image generation is an interesting but challenging task. Existing solutions are either unconditional or dependent on coarse conditions (e.g., pose, text), thus lacking explicit geometry and appearance control of body and garment. Sketching offers such editing ability and has been adopted in various sketch-based face generation and editing solutions. However, directly adapting sketch-based face generation to full-body generation often fails to produce high-fidelity and diverse results due to the high complexity and diversity in the pose, body shape, and garment shape and texture. Recent geometrically controllable diffusion-based methods mainly rely on prompts to generate appearance. It is hard to balance the realism and the faithfulness of their results to the sketch when the input is coarse. This work presents Sketch2Human, the first system for controllable full-body human image generation guided by a semantic sketch (for geometry control) and a reference image (for appearance control). Our solution is based on the latent space of StyleGAN-Human with inverted geometry and appearance latent codes as input. Specifically, we present a sketch encoder trained with a large synthetic dataset sampled from StyleGAN-Human's latent space and directly supervised by sketches rather than real images. Considering the entangled information of partial geometry and texture in StyleGAN-Human and the absence of disentangled datasets, we design a novel training scheme that creates geometry-preserved and appearance-transferred training data to tune a generator to achieve disentangled geometry and appearance control. Although our method is trained with synthetic data, it can also handle hand-drawn sketches. Qualitative and quantitative evaluations demonstrate the superior performance of our method to state-of-the-art methods.

Large-scale datasets and deep generative models have enabled impressive progress in human face reenactment. Existing solutions for face reenactment have focused on processing real face images through facial landmarks by generative models. Different from real human faces, artistic human faces (e.g., those in paintings, cartoons, etc.) often involve exaggerated shapes and various textures. Therefore, directly applying existing solutions to artistic faces often fails to preserve the characteristics of the original artistic faces (e.g., face identity and decorative lines along face contours) due to the domain gap between real and artistic faces. To address these issues, we present ReenactArtFace, the first effective solution for transferring the poses and expressions from human videos to various artistic face images. We achieve artistic face reenactment in a coarse-to-fine manner. First, we perform 3D artistic face reconstruction, which reconstructs a textured 3D artistic face through a 3D morphable model (3DMM) and a 2D parsing map from an input artistic image. The 3DMM can not only rig the expressions better than facial landmarks but also render images under different poses/expressions as coarse reenactment results robustly. However, these coarse results suffer from self-occlusions and lack contour lines. Second, we thus perform artistic face refinement by using a personalized conditional adversarial generative model (cGAN) fine-tuned on the input artistic image and the coarse reenactment results. For high-quality refinement, we propose a contour loss to supervise the cGAN to faithfully synthesize contour lines. Quantitative and qualitative experiments demonstrate that our method achieves better results than the existing solutions.

Face reenactment and reconstruction benefit various applications in self-media, VR, etc. Recent face reenactment methods use 2D facial landmarks to implicitly retarget facial expressions and poses from driving videos to source images, while they suffer from pose and expression preservation issues for cross-identity scenarios, i.e., when the source and the driving subjects are different. Current self-supervised face reconstruction methods also demonstrate impressive results. However, these methods do not handle large expressions well, since their training data lacks samples of large expressions, and 2D facial attributes are inaccurate on such samples. To mitigate the above problems, we propose to explore the inner connection between the two tasks, i.e., using face reconstruction to provide sufficient 3D information for reenactment, and synthesizing videos paired with captured face model parameters through face reenactment to enhance the expression module of face reconstruction. In particular, we propose a novel cascade framework named JR2Net for Joint Face Reconstruction and Reenactment, which begins with the training of a coarse reconstruction network, followed by a 3D-aware face reenactment network based on the coarse reconstruction results. In the end, we train an expression tracking network based on our synthesized videos composed by image-face model parameter pairs. Such an expression tracking network can further enhance the coarse face reconstruction. Extensive experiments show that our JR2Net outperforms the state-of-the-art methods on several face reconstruction and reenactment benchmarks.

In this paper, we propose a 3D information-based facial expression recognition network (3D-FERNet), which effectively combines identity, 3D and 2D expression information for facial expression recognition (FER). The 3DFERNet model consists of three feature encoders to extract identity, 2d expression, and 3d expression features, and one semantic-based feature fusion module to combine these features for further classification. Firstly, the identity encoder is constructed by a face identity recognition network, which is used to perform identity-conditional FER and remove identity bias of FER prediction results. Secondly, we design the 3D expression encoder by a 3D face reconstruction network, which is trained under a semi-supervised method with a variety of unlabeled expression data. The generated 3D features can capture the subtle differences between similar expression cases. Thirdly, for the 2D expression encoder, we can leverage most existing FER networks as the backbone. Finally, after obtaining these facial features, we propose a semantic-based feature fusion module, which is based on the attention mechanism for feature combination. Experiments show that the proposed 3D-FERNet achieves state-of-the-art classification accuracy in multiple benchmark data sets. Moreover, the identity feature encoder and 3D feature encoder can serve as two general modules to plug in most existing FER models.

In recent years, sparse voxel-based methods have be-come the state-of-the-arts for 3D semantic segmentation of indoor scenes, thanks to the powerful 3D CNNs. Nevertheless, being oblivious to the underlying geometry, voxel-based methods suffer from ambiguous features on spatially close objects and struggle with handling complex and irregular geometries due to the lack of geodesic information. In view of this, we present Voxel-Mesh Network (VMNet), a novel 3D deep architecture that operates on the voxel and mesh representations leveraging both the Euclidean and geodesic information. Intuitively, the Euclidean information extracted from voxels can offer contextual cues representing interactions between nearby objects, while the geodesic information extracted from meshes can help separate objects that are spatially close but have disconnected surfaces. To incorporate such information from the two domains, we design an intra-domain attentive module for effective feature aggregation and an inter-domain attentive module for adaptive feature fusion. Experimental results validate the effectiveness of VMNet: specifically, on the challenging ScanNet dataset for large-scale segmentation of indoor scenes, it outperforms the state-of-the-art SparseConvNet and MinkowskiNet (74.6% vs 72.5% and 73.6% in mIoU) with a simpler network structure (17M vs 30M and 38M parameters). Code release: https://github.com/hzykent/VMNet

In this paper, we present a joint multi-task learning framework for semantic segmentation and boundary detection. The critical component in the framework is the iterative pyramid context module (PCM), which couples two tasks and stores the shared latent semantics to interact between the two tasks. For semantic boundary detection, we propose the novel spatial gradient fusion to suppress non-semantic edges. As semantic boundary detection is the dual task of semantic segmentation, we introduce a loss function with boundary consistency constraint to improve the boundary pixel accuracy for semantic segmentation. Our extensive experiments demonstrate superior performance over state-of-the-art works, not only in semantic segmentation but also in semantic boundary detection. In particular, a mean IoU score of 81.8% on Cityscapes test set is achieved without using coarse data or any external data for semantic segmentation. For semantic boundary detection, we improve over previous state-of-the-art works by 9.9% in terms of AP and 6.8% in terms of MF(ODS).

Regression based methods are not performing as well as detection based methods for human pose estimation. A central problem is that the structural information in the pose is not well exploited in the previous regression methods. In this work, we propose a structure-aware regression approach. It adopts a reparameterized pose representation using bones instead of joints. It exploits the joint connection structure to define a compositional loss function that encodes the long range interactions in the pose. It is simple, effective, and general for both 2D and 3D pose estimation in a unified setting. Comprehensive evaluation validates the effectiveness of our approach. It significantly advances the state-of-the-art on Human3.6M [20] and is competitive with state-of-the-art results on MPII [3].