Divya Kothandaraman

Divya Kothandaraman, Sumit Shekhar, Abhilasha Sancheti et al.

We present a novel method, SALAD, for the challenging vision task of adapting a pre-trained "source" domain network to a "target" domain, with a small budget for annotation in the "target" domain and a shift in the label space. Further, the task assumes that the source data is not available for adaptation, due to privacy concerns or otherwise. We postulate that such systems need to jointly optimize the dual task of (i) selecting fixed number of samples from the target domain for annotation and (ii) transfer of knowledge from the pre-trained network to the target domain. To do this, SALAD consists of a novel Guided Attention Transfer Network (GATN) and an active learning function, HAL. The GATN enables feature distillation from pre-trained network to the target network, complemented with the target samples mined by HAL using transfer-ability and uncertainty criteria. SALAD has three key benefits: (i) it is task-agnostic, and can be applied across various visual tasks such as classification, segmentation and detection; (ii) it can handle shifts in output label space from the pre-trained source network to the target domain; (iii) it does not require access to source data for adaptation. We conduct extensive experiments across 3 visual tasks, viz. digits classification (MNIST, SVHN, VISDA), synthetic (GTA5) to real (CityScapes) image segmentation, and document layout detection (PubLayNet to DSSE). We show that our source-free approach, SALAD, results in an improvement of 0.5%-31.3%(across datasets and tasks) over prior adaptation methods that assume access to large amounts of annotated source data for adaptation.

Divya Kothandaraman, Tianrui Guan, Xijun Wang et al.

We present an algorithm, Fourier Activity Recognition (FAR), for UAV video activity recognition. Our formulation uses a novel Fourier object disentanglement method to innately separate out the human agent (which is typically small) from the background. Our disentanglement technique operates in the frequency domain to characterize the extent of temporal change of spatial pixels, and exploits convolution-multiplication properties of Fourier transform to map this representation to the corresponding object-background entangled features obtained from the network. To encapsulate contextual information and long-range space-time dependencies, we present a novel Fourier Attention algorithm, which emulates the benefits of self-attention by modeling the weighted outer product in the frequency domain. Our Fourier attention formulation uses much fewer computations than self-attention. We have evaluated our approach on multiple UAV datasets including UAV Human RGB, UAV Human Night, Drone Action, and NEC Drone. We demonstrate a relative improvement of 8.02% - 38.69% in top-1 accuracy and up to 3 times faster over prior works.

Ruiqi Xian, Xijun Wang, Divya Kothandaraman et al.

We present a new algorithm for selection of informative frames in video action recognition. Our approach is designed for aerial videos captured using a moving camera where human actors occupy a small spatial resolution of video frames. Our algorithm utilizes the motion bias within aerial videos, which enables the selection of motion-salient frames. We introduce the concept of patch mutual information (PMI) score to quantify the motion bias between adjacent frames, by measuring the similarity of patches. We use this score to assess the amount of discriminative motion information contained in one frame relative to another. We present an adaptive frame selection strategy using shifted leaky ReLu and cumulative distribution function, which ensures that the sampled frames comprehensively cover all the essential segments with high motion salience. Our approach can be integrated with any action recognition model to enhance its accuracy. In practice, our method achieves a relative improvement of 2.2 - 13.8% in top-1 accuracy on UAV-Human, 6.8% on NEC Drone, and 9.0% on Diving48 datasets.

James F. Mullen, Divya Kothandaraman, Aniket Bera et al.

We present a novel method for placing a 3D human animation into a 3D scene while maintaining any human-scene interactions in the animation. We use the notion of computing the most important meshes in the animation for the interaction with the scene, which we call "keyframes." These keyframes allow us to better optimize the placement of the animation into the scene such that interactions in the animations (standing, laying, sitting, etc.) match the affordances of the scene (e.g., standing on the floor or laying in a bed). We compare our method, which we call PAAK, with prior approaches, including POSA, PROX ground truth, and a motion synthesis method, and highlight the benefits of our method with a perceptual study. Human raters preferred our PAAK method over the PROX ground truth data 64.6\% of the time. Additionally, in direct comparisons, the raters preferred PAAK over competing methods including 61.5\% compared to POSA.

Divya Kothandaraman, Ming Lin, Dinesh Manocha

We present a learning algorithm for human activity recognition in videos. Our approach is designed for UAV videos, which are mainly acquired from obliquely placed dynamic cameras that contain a human actor along with background motion. Typically, the human actors occupy less than one-tenth of the spatial resolution. Our approach simultaneously harnesses the benefits of frequency domain representations, a classical analysis tool in signal processing, and data driven neural networks. We build a differentiable static-dynamic frequency mask prior to model the salient static and dynamic pixels in the video, crucial for the underlying task of action recognition. We use this differentiable mask prior to enable the neural network to intrinsically learn disentangled feature representations via an identity loss function. Our formulation empowers the network to inherently compute disentangled salient features within its layers. Further, we propose a cost-function encapsulating temporal relevance and spatial content to sample the most important frame within uniformly spaced video segments. We conduct extensive experiments on the UAV Human dataset and the NEC Drone dataset and demonstrate relative improvements of 5.72% - 13.00% over the state-of-the-art and 14.28% - 38.05% over the corresponding baseline model.

Divya Kothandaraman, Kuldeep Kulkarni, Sumit Shekhar et al.

We present ImPoster, a novel algorithm for generating a target image of a 'source' subject performing a 'driving' action. The inputs to our algorithm are a single pair of a source image with the subject that we wish to edit and a driving image with a subject of an arbitrary class performing the driving action, along with the text descriptions of the two images. Our approach is completely unsupervised and does not require any access to additional annotations like keypoints or pose. Our approach builds on a pretrained text-to-image latent diffusion model and learns the characteristics of the source and the driving image by finetuning the diffusion model for a small number of iterations. At inference time, ImPoster performs step-wise text prompting i.e. it denoises by first moving in the direction of the image manifold corresponding to the driving image followed by the direction of the image manifold corresponding to the text description of the desired target image. We propose a novel diffusion guidance formulation, image frequency guidance, to steer the generation towards the manifold of the source subject and the driving action at every step of the inference denoising. Our frequency guidance formulations are derived from the frequency domain properties of images. We extensively evaluate ImPoster on a diverse set of source-driving image pairs to demonstrate improvements over baselines. To the best of our knowledge, ImPoster is the first approach towards achieving both subject-driven as well as action-driven image personalization. Code and data is available at https://github.com/divyakraman/ImPosterDiffusion2024.

Divya Kothandaraman, Tianyi Zhou, Ming Lin et al.

We present HawkI, for synthesizing aerial-view images from text and an exemplar image, without any additional multi-view or 3D information for finetuning or at inference. HawkI uses techniques from classical computer vision and information theory. It seamlessly blends the visual features from the input image within a pretrained text-to-2Dimage stable diffusion model with a test-time optimization process for a careful bias-variance trade-off, which uses an Inverse Perspective Mapping (IPM) homography transformation to provide subtle cues for aerialview synthesis. At inference, HawkI employs a unique mutual information guidance formulation to steer the generated image towards faithfully replicating the semantic details of the input-image, while maintaining a realistic aerial perspective. Mutual information guidance maximizes the semantic consistency between the generated image and the input image, without enforcing pixel-level correspondence between vastly different viewpoints. Through extensive qualitative and quantitative comparisons against text + exemplar-image based methods and 3D/ multi-view based novel-view synthesis methods on proposed synthetic and real datasets, we demonstrate that our method achieves a significantly better bias-variance trade-off towards generating high fidelity aerial-view images.Code and data is available at https://github.com/divyakraman/HawkI2024.

Divya Kothandaraman, Tianyi Zhou, Ming Lin et al.

We present a novel method, Aerial Diffusion, for generating aerial views from a single ground-view image using text guidance. Aerial Diffusion leverages a pretrained text-image diffusion model for prior knowledge. We address two main challenges corresponding to domain gap between the ground-view and the aerial view and the two views being far apart in the text-image embedding manifold. Our approach uses a homography inspired by inverse perspective mapping prior to finetuning the pretrained diffusion model. Additionally, using the text corresponding to the ground-view to finetune the model helps us capture the details in the ground-view image at a relatively low bias towards the ground-view image. Aerial Diffusion uses an alternating sampling strategy to compute the optimal solution on complex high-dimensional manifold and generate a high-fidelity (w.r.t. ground view) aerial image. We demonstrate the quality and versatility of Aerial Diffusion on a plethora of images from various domains including nature, human actions, indoor scenes, etc. We qualitatively prove the effectiveness of our method with extensive ablations and comparisons. To the best of our knowledge, Aerial Diffusion is the first approach that performs ground-to-aerial translation in an unsupervised manner.

Lingdong Wang, Guan-Ming Su, Divya Kothandaraman et al.

Traditional video codecs optimized for pixel fidelity collapse at ultra-low bitrates and produce severe artifacts. This failure arises from a fundamental misalignment between pixel accuracy and human perception. We propose a semantic video compression framework named DiSCo that transmits only the most meaningful information while relying on generative priors for detail synthesis. The source video is decomposed into three compact modalities: a textual description, a spatiotemporally degraded video, and optional sketches or poses that respectively capture semantic, appearance, and motion cues. A conditional video diffusion model then reconstructs high-quality, temporally coherent videos from these compact representations. Temporal forward filling, token interleaving, and modality-specific codecs are proposed to improve multimodal generation and modality compactness. Experiments show that our method outperforms baseline semantic and traditional codecs by 2-10X on perceptual metrics at low bitrates.

Taewon Kang, Divya Kothandaraman, Dinesh Manocha et al.

Recent 3D novel view synthesis (NVS) methods often require extensive 3D data for training, and also typically lack generalization beyond the training distribution. Moreover, they tend to be object centric and struggle with complex and intricate scenes. Conversely, 3D-free methods can generate text-controlled views of complex, in-the-wild scenes using a pretrained stable diffusion model without the need for a large amount of 3D-based training data, but lack camera control. In this paper, we introduce a method capable of generating camera-controlled viewpoints from a single input image, by combining the benefits of 3D-free and 3D-based approaches. Our method excels in handling complex and diverse scenes without extensive training or additional 3D and multiview data. It leverages widely available pretrained NVS models for weak guidance, integrating this knowledge into a 3D-free view synthesis style approach, along with enriching the CLIP vision-language space with 3D camera angle information, to achieve the desired results. Experimental results demonstrate that our method outperforms existing models in both qualitative and quantitative evaluations, achieving high-fidelity, consistent novel view synthesis at desired camera angles across a wide variety of scenes while maintaining accurate, natural detail representation and image clarity across various viewpoints. We also support our method with a comprehensive analysis of 2D image generation models and the 3D space, providing a solid foundation and rationale for our solution.

Divya Kothandaraman, Ming Lin, Dinesh Manocha

We introduce a novel approach for concept blending in pretrained text-to-image diffusion models, aiming to generate images at the intersection of multiple text prompts. At each time step during diffusion denoising, our algorithm forecasts predictions w.r.t. the generated image and makes informed text conditioning decisions. Central to our method is the unique analogy between diffusion models, which are rooted in non-equilibrium thermodynamics, and the Black-Scholes model for financial option pricing. By drawing parallels between key variables in both domains, we derive a robust algorithm for concept blending that capitalizes on the Markovian dynamics of the Black-Scholes framework. Our text-based concept blending algorithm is data-efficient, meaning it does not need additional training. Furthermore, it operates without human intervention or hyperparameter tuning. We highlight the benefits of our approach by comparing it qualitatively and quantitatively to other text based concept blending techniques, including linear interpolation, alternating prompts, step-wise prompt switching, and CLIP-guided prompt selection across various scenarios such as single object per text prompt, multiple objects per text prompt and objects against backgrounds. Our work shows that financially inspired techniques can enhance text-to-image concept blending in generative AI, paving the way for broader innovation. Code is available at https://github.com/divyakraman/BlackScholesDiffusion2024.

Divya Kothandaraman, Kihyuk Sohn, Ruben Villegas et al.

We present a method for multi-concept customization of pretrained text-to-video (T2V) models. Intuitively, the multi-concept customized video can be derived from the (non-linear) intersection of the video manifolds of the individual concepts, which is not straightforward to find. We hypothesize that sequential and controlled walking towards the intersection of the video manifolds, directed by text prompting, leads to the solution. To do so, we generate the various concepts and their corresponding interactions, sequentially, in an autoregressive manner. Our method can generate videos of multiple custom concepts (subjects, action and background) such as a teddy bear running towards a brown teapot, a dog playing violin and a teddy bear swimming in the ocean. We quantitatively evaluate our method using videoCLIP and DINO scores, in addition to human evaluation. Videos for results presented in this paper can be found at https://github.com/divyakraman/MultiConceptVideo2024.

Divya Kothandaraman, Jaclyn Pytlarz

Memorization in large-scale text-to-image diffusion models poses significant security and intellectual property risks, enabling adversarial attribute extraction and the unauthorized reproduction of sensitive or proprietary features. While conventional dememorization techniques, such as regularization and data filtering, limit overfitting to specific training examples, they fail to systematically prevent the internalization of prohibited concept-level features. Simply discarding all images containing a sensitive feature wastes invaluable training data, necessitating a method for selective learning at the concept level. We introduce a gradient projection method designed to enforce a stringent requirement of concept-level feature exclusion. Our defense operates during backpropagation by systematically identifying and excising training signals aligned with embeddings of prohibited attributes. Specifically, we project each gradient update onto the orthogonal complement of the sensitive feature's embedding space, thereby zeroing out its influence on the model's weights. Our method integrates seamlessly into standard diffusion model training pipelines and complements existing defenses. We analyze our method against an adversary aiming for feature extraction. In extensive experiments, we demonstrate that our framework drastically reduces memorization while rigorously preserving generation quality and semantic fidelity. By reframing memorization control as selective learning, our approach establishes a new paradigm for IP-safe and privacy-preserving generative AI.

Bowen Chen, Jake Zuena, Alan C. Bovik et al.

Precise spatial control in diffusion-based style transfer remains challenging. This challenge arises because diffusion models treat style as a global feature and lack explicit spatial grounding of style representations, making it difficult to restrict style application to specific objects or regions. To our knowledge, existing diffusion models are unable to perform true localized style transfer, typically relying on handcrafted masks or multi-stage post-processing that introduce boundary artifacts and limit generalization. To address this, we propose an attention-supervised diffusion framework that explicitly teaches the model where to apply a given style by aligning the attention scores of style tokens with object masks during training. Two complementary objectives, a Focus loss based on KL divergence and a Cover loss using binary cross-entropy, jointly encourage accurate localization and dense coverage. A modular LoRA-MoE design further enables efficient and scalable multi-style adaptation. To evaluate localized stylization, we introduce the Regional Style Editing Score, which measures Regional Style Matching through CLIP-based similarity within the target region and Identity Preservation via masked LPIPS and pixel-level consistency on unedited areas. Experiments show that our method achieves mask-free, single-object style transfer at inference, producing regionally accurate and visually coherent results that outperform existing diffusion-based editing approaches.

Taewon Kang, Divya Kothandaraman, Ming C. Lin

Generating coherent long-form video sequences from discrete input using only text prompts is a critical task in content creation. While diffusion-based models excel at short video synthesis, long-form storytelling from text remains largely unexplored and a challenge due to difficulties in temporal coherency, preserving semantic meaning, and maintaining both scene context and action continuity across the video. We introduce a novel storytelling framework that achieves this by integrating scene and action prompts through dynamics-inspired prompt mixing. Specifically, we first present a bidirectional time-weighted latent blending strategy to ensure temporal consistency between segments of the long-form video being generated. We then propose a dynamics-informed prompt weighting (DIPW) mechanism that adaptively balances the influence of scene and action prompts at each diffusion timestep by jointly considering CLIP-based alignment, narrative continuity, and temporal smoothness. To further enhance motion continuity, we incorporate a semantic action representation to encode high-level action semantics into the blending process, dynamically adjusting transitions based on action similarity and ensuring smooth yet adaptable motion changes. Latent space blending maintains spatial coherence between objects in a scene, while time-weighted blending enforces bidirectional constraints for temporal consistency. The resulting integrative system prevents abrupt transitions while ensuring fluid storytelling that faithfully reflects both scene and action cues. Extensive experiments demonstrate significant improvements over baselines, achieving temporally consistent and visually compelling video narratives without any additional training. This approach bridges the gap between short clips and extended video to establish a new paradigm in GenAI-driven video synthesis from text.

Pooja Guhan, Divya Kothandaraman, Tsung-Wei Huang et al.

We introduce CamMimic, an innovative algorithm tailored for dynamic video editing needs. It is designed to seamlessly transfer the camera motion observed in a given reference video onto any scene of the user's choice in a zero-shot manner without requiring any additional data. Our algorithm achieves this using a two-phase strategy by leveraging a text-to-video diffusion model. In the first phase, we develop a multi-concept learning method using a combination of LoRA layers and an orthogonality loss to capture and understand the underlying spatial-temporal characteristics of the reference video as well as the spatial features of the user's desired scene. The second phase proposes a unique homography-based refinement strategy to enhance the temporal and spatial alignment of the generated video. We demonstrate the efficacy of our method through experiments conducted on a dataset containing combinations of diverse scenes and reference videos containing a variety of camera motions. In the absence of an established metric for assessing camera motion transfer between unrelated scenes, we propose CameraScore, a novel metric that utilizes homography representations to measure camera motion similarity between the reference and generated videos. Extensive quantitative and qualitative evaluations demonstrate that our approach generates high-quality, motion-enhanced videos. Additionally, a user study reveals that 70.31% of participants preferred our method for scene preservation, while 90.45% favored it for motion transfer. We hope this work lays the foundation for future advancements in camera motion transfer across different scenes.

Tianrui Guan, Divya Kothandaraman, Rohan Chandra et al.

We propose GANav, a novel group-wise attention mechanism to identify safe and navigable regions in off-road terrains and unstructured environments from RGB images. Our approach classifies terrains based on their navigability levels using coarse-grained semantic segmentation. Our novel group-wise attention loss enables any backbone network to explicitly focus on the different groups' features with low spatial resolution. Our design leads to efficient inference while maintaining a high level of accuracy compared to existing SOTA methods. Our extensive evaluations on the RUGD and RELLIS-3D datasets shows that GANav achieves an improvement over the SOTA mIoU by 2.25-39.05% on RUGD and 5.17-19.06% on RELLIS-3D. We interface GANav with a deep reinforcement learning-based navigation algorithm and highlight its benefits in terms of navigation in real-world unstructured terrains. We integrate our GANav-based navigation algorithm with ClearPath Jackal and Husky robots, and observe an increase of 10% in terms of success rate, 2-47% in terms of selecting the surface with the best navigability and a decrease of 4.6-13.9% in trajectory roughness. Further, GANav reduces the false positive rate of forbidden regions by 37.79%. Code, videos, and a full technical report are available at https://gamma.umd.edu/offroad/.

Divya Kothandaraman, Rohan Chandra, Dinesh Manocha

We present a novel approach for unsupervised road segmentation in adverse weather conditions such as rain or fog. This includes a new algorithm for source-free domain adaptation (SFDA) using self-supervised learning. Moreover, our approach uses several techniques to address various challenges in SFDA and improve performance, including online generation of pseudo-labels and self-attention as well as use of curriculum learning, entropy minimization and model distillation. We have evaluated the performance on $6$ datasets corresponding to real and synthetic adverse weather conditions. Our method outperforms all prior works on unsupervised road segmentation and SFDA by at least 10.26%, and improves the training time by 18-180x. Moreover, our self-supervised algorithm exhibits similar accuracy performance in terms of mIOU score as compared to prior supervised methods.

Divya Kothandaraman, Athira Nambiar, Anurag Mittal

Practical autonomous driving systems face two crucial challenges: memory constraints and domain gap issues. In this paper, we present a novel approach to learn domain adaptive knowledge in models with limited memory, thus bestowing the model with the ability to deal with these issues in a comprehensive manner. We term this as "Domain Adaptive Knowledge Distillation" and address the same in the context of unsupervised domain-adaptive semantic segmentation by proposing a multi-level distillation strategy to effectively distil knowledge at different levels. Further, we introduce a novel cross entropy loss that leverages pseudo labels from the teacher. These pseudo teacher labels play a multifaceted role towards: (i) knowledge distillation from the teacher network to the student network & (ii) serving as a proxy for the ground truth for target domain images, where the problem is completely unsupervised. We introduce four paradigms for distilling domain adaptive knowledge and carry out extensive experiments and ablation studies on real-to-real as well as synthetic-to-real scenarios. Our experiments demonstrate the profound success of our proposed method.

Divya Kothandaraman, Rohan Chandra, Dinesh Manocha

We present an unsupervised adaptation approach for visual scene understanding in unstructured traffic environments. Our method is designed for unstructured real-world scenarios with dense and heterogeneous traffic consisting of cars, trucks, two-and three-wheelers, and pedestrians. We describe a new semantic segmentation technique based on unsupervised domain adaptation (DA), that can identify the class or category of each region in RGB images or videos. We also present a novel self-training algorithm (Alt-Inc) for multi-source DA that improves the accuracy. Our overall approach is a deep learning-based technique and consists of an unsupervised neural network that achieves 87.18% accuracy on the challenging India Driving Dataset. Our method works well on roads that may not be well-marked or may include dirt, unidentifiable debris, potholes, etc. A key aspect of our approach is that it can also identify objects that are encountered by the model for the fist time during the testing phase. We compare our method against the state-of-the-art methods and show an improvement of 5.17% - 42.9%. Furthermore, we also conduct user studies that qualitatively validate the improvements in visual scene understanding of unstructured driving environments.

Yuqian Zhou, Michael Kwan, Kyle Tolentino et al.

This paper is the report of the first Under-Display Camera (UDC) image restoration challenge in conjunction with the RLQ workshop at ECCV 2020. The challenge is based on a newly-collected database of Under-Display Camera. The challenge tracks correspond to two types of display: a 4k Transparent OLED (T-OLED) and a phone Pentile OLED (P-OLED). Along with about 150 teams registered the challenge, eight and nine teams submitted the results during the testing phase for each track. The results in the paper are state-of-the-art restoration performance of Under-Display Camera Restoration. Datasets and paper are available at https://yzhouas.github.io/projects/UDC/udc.html.

Varun Sundar, Sumanth Hegde, Divya Kothandaraman et al.

Under Display Cameras present a promising opportunity for phone manufacturers to achieve bezel-free displays by positioning the camera behind semi-transparent OLED screens. Unfortunately, such imaging systems suffer from severe image degradation due to light attenuation and diffraction effects. In this work, we present Deep Atrous Guided Filter (DAGF), a two-stage, end-to-end approach for image restoration in UDC systems. A Low-Resolution Network first restores image quality at low-resolution, which is subsequently used by the Guided Filter Network as a filtering input to produce a high-resolution output. Besides the initial downsampling, our low-resolution network uses multiple, parallel atrous convolutions to preserve spatial resolution and emulates multi-scale processing. Our approach's ability to directly train on megapixel images results in significant performance improvement. We additionally propose a simple simulation scheme to pre-train our model and boost performance. Our overall framework ranks 2nd and 5th in the RLQ-TOD'20 UDC Challenge for POLED and TOLED displays, respectively.