Fuchen Long
Research Subjects
Fuchen Long has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Zhaofan Qiu
HiDream.ai Inc., Beijing, China
11 shared publications - Ting Yao
HiDream.ai Inc., Beijing, China
10 shared publications - Tao Mei
HiDream.ai Inc., Beijing, China
9 shared publications - Jiebo Luo
University of Rochester, Rochester, New York, USA
5 shared publications - Yan Shu
University of Science and Technology of China
3 shared publications - Zhikai Chen
MoE Key Laboratory of Brain-inspired Intelligent Perception and Cognition, University of Science and Technology of China, Hefei, China
3 shared publications
Latest Publications
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28, 13504-13517
Watch, Localize and Answer: Actom-Centric Video Large Language Models
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28, 13504-13517
Watch, Localize and Answer: Actom-Centric Video Large Language Models
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 13737-13742
Identity-Preserving Video Generation Challenge
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 13737-13742
Open AccessIdentity-Preserving Video Generation Challenge
- 6 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 42-56
Tuning-Free High-Resolution Video Diffusion With Spatial-Temporal Latent Grouping
Recent advances in text-to-video generation have demonstrated the substantial superiority of diffusion models. Nevertheless, generating high-resolution videos based on text description still faces a great challenge due to the enormous computation overhead for video diffusion model training. In this paper, we present a tuning-free video diffusion approach with Spatial-Temporal LAtent Grouping (ST-LAG), for highresolution video generation. ST-LAG exploits the prior knowledge of a pre-trained low-resolution video diffusion model for regionwise video latent denoising, and then combines all the denoised regions of video latent as a whole one to achieve global-wise spatial-temporal coherence. Specifically, ST-LAG denoises the whole video latents via two deliberately designed modules, e.g., Spatial Latent Grouping (SLG) and Temporal Latent Grouping (TLG), at spatial and temporal level, respectively. SLG spatially slices the latent of each frame into different local patches, and then feeds them into the low-resolution video diffusion model for local-region latent denoising. A text re-weighting scheme is devised in SLG to strength the cross-attention between features of text tokens and spatial regions to facilitate spatial-level finegrained details generation. TLG capitalizes on the segmentlevel latent grouping to match the length of each denoised local segment with the frame number in the training stage. The well aligned temporal receptive field facilitates better preservation of motion patterns. In each denoising step, all groups of video latent at spatial and temporal levels are fused together for highresolution video generation. Extensive experiments conducted on the ECTV-Prompt dataset demonstrate the effectiveness of our approach quantitatively and qualitatively.
- 6 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 42-56
Tuning-Free High-Resolution Video Diffusion With Spatial-Temporal Latent Grouping
Recent advances in text-to-video generation have demonstrated the substantial superiority of diffusion models. Nevertheless, generating high-resolution videos based on text description still faces a great challenge due to the enormous computation overhead for video diffusion model training. In this paper, we present a tuning-free video diffusion approach with Spatial-Temporal LAtent Grouping (ST-LAG), for highresolution video generation. ST-LAG exploits the prior knowledge of a pre-trained low-resolution video diffusion model for regionwise video latent denoising, and then combines all the denoised regions of video latent as a whole one to achieve global-wise spatial-temporal coherence. Specifically, ST-LAG denoises the whole video latents via two deliberately designed modules, e.g., Spatial Latent Grouping (SLG) and Temporal Latent Grouping (TLG), at spatial and temporal level, respectively. SLG spatially slices the latent of each frame into different local patches, and then feeds them into the low-resolution video diffusion model for local-region latent denoising. A text re-weighting scheme is devised in SLG to strength the cross-attention between features of text tokens and spatial regions to facilitate spatial-level finegrained details generation. TLG capitalizes on the segmentlevel latent grouping to match the length of each denoised local segment with the frame number in the training stage. The well aligned temporal receptive field facilitates better preservation of motion patterns. In each denoising step, all groups of video latent at spatial and temporal levels are fused together for highresolution video generation. Extensive experiments conducted on the ECTV-Prompt dataset demonstrate the effectiveness of our approach quantitatively and qualitatively.
- 11 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 26 (15209210) , 8915-8927
Learning Temporal Dynamics in Videos With Image Transformer
Temporal dynamics represent the evolving of video content over time, which are critical for action recognition. In this paper, we ask the question: can the off-the-shelf image transformer architecture learn temporal dynamics in videos? To this end, we propose Multidimensional Stacked Image (MSImage) as a new arrangement of video data, which can be fed to image transformers. Technically, MSImage is a high-resolution image that is composed of several evenly-sampled video clips stacked along the channel and space dimensions. The frames in each clip are concatenated along the channel dimension for the transformers to infer short-term dynamics. Meanwhile, the clips are arranged on different spatial positions for learning long-term dynamics. On this basis we propose MSImageFormer-a new variant of image transformer that takes MSImage as the input and is jointly optimized by video classification loss and a new dynamics enhancement loss. The network optimization attends to the high-frequency component of MSImage, avoiding overfitting to static visual patterns. We empirically demonstrate the merits of the MSImageFormer on six action recognition benchmarks. With only 2D image transformer as the classifier, our MSImageFormer achieves 85.3% and 69.7% top-1 accuracy on Kinetics-400 and Something-Something V2 datasets, respectively. Despite with fewer computations, the results are comparable to the SOTA 3D CNNs and video transformers.
- 11 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 26 (15209210) , 8915-8927
Learning Temporal Dynamics in Videos With Image Transformer
Temporal dynamics represent the evolving of video content over time, which are critical for action recognition. In this paper, we ask the question: can the off-the-shelf image transformer architecture learn temporal dynamics in videos? To this end, we propose Multidimensional Stacked Image (MSImage) as a new arrangement of video data, which can be fed to image transformers. Technically, MSImage is a high-resolution image that is composed of several evenly-sampled video clips stacked along the channel and space dimensions. The frames in each clip are concatenated along the channel dimension for the transformers to infer short-term dynamics. Meanwhile, the clips are arranged on different spatial positions for learning long-term dynamics. On this basis we propose MSImageFormer-a new variant of image transformer that takes MSImage as the input and is jointly optimized by video classification loss and a new dynamics enhancement loss. The network optimization attends to the high-frequency component of MSImage, avoiding overfitting to static visual patterns. We empirically demonstrate the merits of the MSImageFormer on six action recognition benchmarks. With only 2D image transformer as the classifier, our MSImageFormer achieves 85.3% and 69.7% top-1 accuracy on Kinetics-400 and Something-Something V2 datasets, respectively. Despite with fewer computations, the results are comparable to the SOTA 3D CNNs and video transformers.
- 26 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 26 (15209210) , 8717-8729
Learning 3D Shape Latent for Point Cloud Completion
By formulating the data generation as a sequence procedure of denoising autoencoding, diffusion models have achieved superior in-painting performance on image data and beyond. Nevertheless, it is not trivial when capitalizing on diffusion models to generate missing 3D points. The difficulty originates from the intrinsic structure where 3D point cloud is a set of unordered and irregular coordinates. That motivates us to delve into the 3D structural information for designing point cloud encoder-decoder and shape latent generator, to precisely formulate the latent distribution of the complete point cloud and partial observation. In this paper, we propose Point cloud completion with Latent Diffusion Models (PointLDM), a new approach that leverages the conditional denoising diffusion probabilistic modeling (DDPM) in the 3D latent space for shape reconstruction. The architecture of PointLDM consists of a transformer-based variational auto-encoder (VAE) to model the complete shape latent, and a diffusion network for shape latent prediction. The encoder of VAE exploits both of global shape latent and local point features in shape distribution learning. With the learnt shape latent, the decoder first decodes the shape latent into coarse points, and then recovers the fine-grained details around each coarse point by deforming a 2D grid. To reconstruct the shape latent from partial observation, the diffusion network treats the partial observation as the conditional input and generates the shape latent via DDPM. Extensive experiments conducted on MVP, Completion3D, and KITTI quantitatively and qualitatively demonstrate the efficacy of PointLDM over the state-of-the-art shape completion approaches.
- 26 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 26 (15209210) , 8717-8729
Learning 3D Shape Latent for Point Cloud Completion
By formulating the data generation as a sequence procedure of denoising autoencoding, diffusion models have achieved superior in-painting performance on image data and beyond. Nevertheless, it is not trivial when capitalizing on diffusion models to generate missing 3D points. The difficulty originates from the intrinsic structure where 3D point cloud is a set of unordered and irregular coordinates. That motivates us to delve into the 3D structural information for designing point cloud encoder-decoder and shape latent generator, to precisely formulate the latent distribution of the complete point cloud and partial observation. In this paper, we propose Point cloud completion with Latent Diffusion Models (PointLDM), a new approach that leverages the conditional denoising diffusion probabilistic modeling (DDPM) in the 3D latent space for shape reconstruction. The architecture of PointLDM consists of a transformer-based variational auto-encoder (VAE) to model the complete shape latent, and a diffusion network for shape latent prediction. The encoder of VAE exploits both of global shape latent and local point features in shape distribution learning. With the learnt shape latent, the decoder first decodes the shape latent into coarse points, and then recovers the fine-grained details around each coarse point by deforming a 2D grid. To reconstruct the shape latent from partial observation, the diffusion network treats the partial observation as the conditional input and generates the shape latent via DDPM. Extensive experiments conducted on MVP, Completion3D, and KITTI quantitatively and qualitatively demonstrate the efficacy of PointLDM over the state-of-the-art shape completion approaches.
- 4 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 33 (10577149) , 625-638
A Closer Look at the Reflection Formulation in Single Image Reflection Removal
How to model the effect of reflection is crucial for single image reflection removal (SIRR) task. Modern SIRR methods usually simplify the reflection formulation with the assumption of linear combination of a transmission layer and a reflection layer. However, the large variations in image content and the real-world picture-taking conditions often result in far more complex reflection. In this paper, we introduce a new screen-blur combination based on two important factors, namely the intensity and the blurriness of reflection, to better characterize the reflection formulation in SIRR. Specifically, we present Screen-blur Reflection Networks (SRNet), which executes the screen-blur formulation in its network design and adapts to the complex reflection on real scenes. Technically, SRNet consists of three components: a blended image generator, a reflection estimator and a reflection removal module. The image generator exploits the screen-blur combination to synthesize the training blended images. The reflection estimator learns the reflection layer and a blur degree that measures the level of blurriness for reflection. The reflection removal module further uses the blended image, blur degree and reflection layer to filter out the transmission layer in a cascaded manner. Superior results on three different SIRR methods are reported when generating the training data on the principle of the screen-blur combination. Moreover, extensive experiments on six datasets quantitatively and qualitatively demonstrate the efficacy of SRNet over the state-of-the-art methods.
- 4 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 33 (10577149) , 625-638
A Closer Look at the Reflection Formulation in Single Image Reflection Removal
How to model the effect of reflection is crucial for single image reflection removal (SIRR) task. Modern SIRR methods usually simplify the reflection formulation with the assumption of linear combination of a transmission layer and a reflection layer. However, the large variations in image content and the real-world picture-taking conditions often result in far more complex reflection. In this paper, we introduce a new screen-blur combination based on two important factors, namely the intensity and the blurriness of reflection, to better characterize the reflection formulation in SIRR. Specifically, we present Screen-blur Reflection Networks (SRNet), which executes the screen-blur formulation in its network design and adapts to the complex reflection on real scenes. Technically, SRNet consists of three components: a blended image generator, a reflection estimator and a reflection removal module. The image generator exploits the screen-blur combination to synthesize the training blended images. The reflection estimator learns the reflection layer and a blur degree that measures the level of blurriness for reflection. The reflection removal module further uses the blended image, blur degree and reflection layer to filter out the transmission layer in a cascaded manner. Superior results on three different SIRR methods are reported when generating the training data on the principle of the screen-blur combination. Moreover, extensive experiments on six datasets quantitatively and qualitatively demonstrate the efficacy of SRNet over the state-of-the-art methods.
- 30 July 2023
- Published by Springer Nature
- p. 302-312
Involving Density Prior for 3D Point Cloud Contrastive Learning
- 30 July 2023
- Published by Springer Nature
- p. 302-312
Involving Density Prior for 3D Point Cloud Contrastive Learning
- 31 March 2023
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 131 (7) , 1704-1721
Bi-calibration Networks for Weakly-Supervised Video Representation Learning
- 31 March 2023
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 131 (7) , 1704-1721
Bi-calibration Networks for Weakly-Supervised Video Representation Learning