Zhaoyang Huang

Rongyao Fang, Shilin Yan, Zhaoyang Huang et al.

Empowering models to dynamically accomplish tasks specified through natural language instructions represents a promising path toward more capable and general artificial intelligence. In this work, we introduce InstructSeq, an instruction-conditioned multi-modal modeling framework that unifies diverse vision tasks through flexible natural language control and handling of both visual and textual data. InstructSeq employs a multimodal transformer architecture encompassing visual, language, and sequential modeling. We utilize a visual encoder to extract image features and a text encoder to encode instructions. An autoregressive transformer fuses the representations and generates sequential task outputs. By training with LLM-generated natural language instructions, InstructSeq acquires a strong comprehension of free-form instructions for specifying visual tasks. This provides an intuitive interface for directing capabilities using flexible natural instructions. Without any task-specific tuning, InstructSeq achieves compelling performance on semantic segmentation, referring expression segmentation/comprehension, and image captioning. The flexible control and multi-task unification empower the model with more human-like versatility and generalizability for computer vision. The code will be released soon at https://github.com/rongyaofang/InstructSeq.

Yijin Li, Zhaoyang Huang, Shuo Chen et al.

Event cameras provide high temporal precision, low data rates, and high dynamic range visual perception, which are well-suited for optical flow estimation. While data-driven optical flow estimation has obtained great success in RGB cameras, its generalization performance is seriously hindered in event cameras mainly due to the limited and biased training data. In this paper, we present a novel simulator, BlinkSim, for the fast generation of large-scale data for event-based optical flow. BlinkSim incorporates a configurable rendering engine alongside an event simulation suite. By leveraging the wealth of current 3D assets, the rendering engine enables us to automatically build up thousands of scenes with different objects, textures, and motion patterns and render very high-frequency images for realistic event data simulation. Based on BlinkSim, we construct a large training dataset and evaluation benchmark BlinkFlow that contains sufficient, diversiform, and challenging event data with optical flow ground truth. Experiments show that BlinkFlow improves the generalization performance of state-of-the-art methods by more than 40\% on average and up to 90\%. Moreover, we further propose an Event-based optical Flow transFormer (E-FlowFormer) architecture. Powered by our BlinkFlow, E-FlowFormer outperforms the SOTA methods by up to 91\% on the MVSEC dataset and 14\% on the DSEC dataset and presents the best generalization performance. The source code and data are available at https://zju3dv.github.io/blinkflow/.

Yichen Shen, Yijin Li, Shuo Chen et al.

Event cameras, known for their high temporal resolution and ability to capture asynchronous changes, have gained significant attention for their potential in feature tracking, especially in challenging conditions. However, event cameras lack the fine-grained texture information that conventional cameras provide, leading to error accumulation in tracking. To address this, we propose a novel framework, BlinkTrack, which integrates event data with grayscale images for high-frequency feature tracking. Our method extends the traditional Kalman filter into a learning-based framework, utilizing differentiable Kalman filters in both event and image branches. This approach improves single-modality tracking and effectively solves the data association and fusion from asynchronous event and image data. We also introduce new synthetic and augmented datasets to better evaluate our model. Experimental results indicate that BlinkTrack significantly outperforms existing methods, exceeding 80 FPS with multi-modality data and 100 FPS with preprocessed event data. Codes and dataset are available at https://github.com/ColieShen/BlinkTrack.

Yitong Dong, Yijin Li, Zhaoyang Huang et al.

In this paper, we propose a novel multi-view stereo (MVS) framework that gets rid of the depth range prior. Unlike recent prior-free MVS methods that work in a pair-wise manner, our method simultaneously considers all the source images. Specifically, we introduce a Multi-view Disparity Attention (MDA) module to aggregate long-range context information within and across multi-view images. Considering the asymmetry of the epipolar disparity flow, the key to our method lies in accurately modeling multi-view geometric constraints. We integrate pose embedding to encapsulate information such as multi-view camera poses, providing implicit geometric constraints for multi-view disparity feature fusion dominated by attention. Additionally, we construct corresponding hidden states for each source image due to significant differences in the observation quality of the same pixel in the reference frame across multiple source frames. We explicitly estimate the quality of the current pixel corresponding to sampled points on the epipolar line of the source image and dynamically update hidden states through the uncertainty estimation module. Extensive results on the DTU dataset and Tanks&Temple benchmark demonstrate the effectiveness of our method. The code is available at our project page: https://zju3dv.github.io/GD-PoseMVS/.

Zhaoyang Huang, Han Zhou, Yijin Li et al.

Visual localization is of great importance in robotics and computer vision. Recently, scene coordinate regression based methods have shown good performance in visual localization in small static scenes. However, it still estimates camera poses from many inferior scene coordinates. To address this problem, we propose a novel visual localization framework that establishes 2D-to-3D correspondences between the query image and the 3D map with a series of learnable scene-specific landmarks. In the landmark generation stage, the 3D surfaces of the target scene are over-segmented into mosaic patches whose centers are regarded as the scene-specific landmarks. To robustly and accurately recover the scene-specific landmarks, we propose the Voting with Segmentation Network (VS-Net) to segment the pixels into different landmark patches with a segmentation branch and estimate the landmark locations within each patch with a landmark location voting branch. Since the number of landmarks in a scene may reach up to 5000, training a segmentation network with such a large number of classes is both computation and memory costly for the commonly used cross-entropy loss. We propose a novel prototype-based triplet loss with hard negative mining, which is able to train semantic segmentation networks with a large number of labels efficiently. Our proposed VS-Net is extensively tested on multiple public benchmarks and can outperform state-of-the-art visual localization methods. Code and models are available at \href{https://github.com/zju3dv/VS-Net}{https://github.com/zju3dv/VS-Net}.

Fu-Yun Wang, Zhaoyang Huang, Weikang Bian et al.

This paper introduces an effective method for computation-efficient personalized style video generation without requiring access to any personalized video data. It reduces the necessary generation time of similarly sized video diffusion models from 25 seconds to around 1 second while maintaining the same level of performance. The method's effectiveness lies in its dual-level decoupling learning approach: 1) separating the learning of video style from video generation acceleration, which allows for personalized style video generation without any personalized style video data, and 2) separating the acceleration of image generation from the acceleration of video motion generation, enhancing training efficiency and mitigating the negative effects of low-quality video data.

Weikang Bian, Zhaoyang Huang, Xiaoyu Shi et al.

4D video control is essential in video generation as it enables the use of sophisticated lens techniques, such as multi-camera shooting and dolly zoom, which are currently unsupported by existing methods. Training a video Diffusion Transformer (DiT) directly to control 4D content requires expensive multi-view videos. Inspired by Monocular Dynamic novel View Synthesis (MDVS) that optimizes a 4D representation and renders videos according to different 4D elements, such as camera pose and object motion editing, we bring pseudo 4D Gaussian fields to video generation. Specifically, we propose a novel framework that constructs a pseudo 4D Gaussian field with dense 3D point tracking and renders the Gaussian field for all video frames. Then we finetune a pretrained DiT to generate videos following the guidance of the rendered video, dubbed as GS-DiT. To boost the training of the GS-DiT, we also propose an efficient Dense 3D Point Tracking (D3D-PT) method for the pseudo 4D Gaussian field construction. Our D3D-PT outperforms SpatialTracker, the state-of-the-art sparse 3D point tracking method, in accuracy and accelerates the inference speed by two orders of magnitude. During the inference stage, GS-DiT can generate videos with the same dynamic content while adhering to different camera parameters, addressing a significant limitation of current video generation models. GS-DiT demonstrates strong generalization capabilities and extends the 4D controllability of Gaussian splatting to video generation beyond just camera poses. It supports advanced cinematic effects through the manipulation of the Gaussian field and camera intrinsics, making it a powerful tool for creative video production. Demos are available at https://wkbian.github.io/Projects/GS-DiT/.

Junjie Ni, Guofeng Zhang, Guanglin Li et al.

We tackle the efficiency problem of learning local feature matching. Recent advancements have given rise to purely CNN-based and transformer-based approaches, each augmented with deep learning techniques. While CNN-based methods often excel in matching speed, transformer-based methods tend to provide more accurate matches. We propose an efficient transformer-based network architecture for local feature matching. This technique is built on constructing multiple homography hypotheses to approximate the continuous correspondence in the real world and uni-directional cross-attention to accelerate the refinement. On the YFCC100M dataset, our matching accuracy is competitive with LoFTR, a state-of-the-art transformer-based architecture, while the inference speed is boosted to 4 times, even outperforming the CNN-based methods. Comprehensive evaluations on other open datasets such as Megadepth, ScanNet, and HPatches demonstrate our method's efficacy, highlighting its potential to significantly enhance a wide array of downstream applications.

Ruohao Zhan, Yijin Li, Yisheng He et al.

Sketches serve as fundamental blueprints in artistic creation because sketch editing is easier and more intuitive than pixel-level RGB image editing for painting artists, yet sketch generation remains unexplored despite advancements in generative models. We propose a novel framework CoProSketch, providing prominent controllability and details for sketch generation with diffusion models. A straightforward method is fine-tuning a pretrained image generation diffusion model with binarized sketch images. However, we find that the diffusion models fail to generate clear binary images, which makes the produced sketches chaotic. We thus propose to represent the sketches by unsigned distance field (UDF), which is continuous and can be easily decoded to sketches through a lightweight network. With CoProSketch, users generate a rough sketch from a bounding box and a text prompt. The rough sketch can be manually edited and fed back into the model for iterative refinement and will be decoded to a detailed sketch as the final result. Additionally, we curate the first large-scale text-sketch paired dataset as the training data. Experiments demonstrate superior semantic consistency and controllability over baselines, offering a practical solution for integrating user feedback into generative workflows.

Zhaoyang Huang, Xiaoyu Shi, Chao Zhang et al.

We introduce optical Flow transFormer, dubbed as FlowFormer, a transformer-based neural network architecture for learning optical flow. FlowFormer tokenizes the 4D cost volume built from an image pair, encodes the cost tokens into a cost memory with alternate-group transformer (AGT) layers in a novel latent space, and decodes the cost memory via a recurrent transformer decoder with dynamic positional cost queries. On the Sintel benchmark, FlowFormer achieves 1.159 and 2.088 average end-point-error (AEPE) on the clean and final pass, a 16.5% and 15.5% error reduction from the best published result (1.388 and 2.47). Besides, FlowFormer also achieves strong generalization performance. Without being trained on Sintel, FlowFormer achieves 1.01 AEPE on the clean pass of Sintel training set, outperforming the best published result (1.29) by 21.7%.

Yan Xu, Zhaoyang Huang, Kwan-Yee Lin et al.

Recent learning-based LiDAR odometry methods have demonstrated their competitiveness. However, most methods still face two substantial challenges: 1) the 2D projection representation of LiDAR data cannot effectively encode 3D structures from the point clouds; 2) the needs for a large amount of labeled data for training limit the application scope of these methods. In this paper, we propose a self-supervised LiDAR odometry method, dubbed SelfVoxeLO, to tackle these two difficulties. Specifically, we propose a 3D convolution network to process the raw LiDAR data directly, which extracts features that better encode the 3D geometric patterns. To suit our network to self-supervised learning, we design several novel loss functions that utilize the inherent properties of LiDAR point clouds. Moreover, an uncertainty-aware mechanism is incorporated in the loss functions to alleviate the interference of moving objects/noises. We evaluate our method's performances on two large-scale datasets, i.e., KITTI and Apollo-SouthBay. Our method outperforms state-of-the-art unsupervised methods by 27%/32% in terms of translational/rotational errors on the KITTI dataset and also performs well on the Apollo-SouthBay dataset. By including more unlabelled training data, our method can further improve performance comparable to the supervised methods.