MTech Projects for Computer Science — Deep Learning & Digital Forensics 2026
Computer Science MTech projects in 2026 strongly favour Deep Learning (Vision Transformers, Multimodal models, Generative AI) and Digital / Cyber Forensics (memory, network, malware, multimedia forensics and AI-assisted investigation). These domains offer high research value and excellent placement relevance.
Why choose these projects at ProjectsatBangalore?
- IEEE / CVPR / NeurIPS / USENIX style base papers
- Complete PyTorch / TensorFlow code
- Forensic tool pipelines (Autopsy, Volatility, Wireshark)
- Ready datasets or generation scripts
- VTU / Anna / JNTU format reports
- 20–25 slide PPT + 50+ viva Q&A
- Architecture & investigation flow diagrams
- Mentoring via WhatsApp / Zoom
Deep Learning
ViTs, Multimodal LLMs, Diffusion models, Efficient transformers
Digital Forensics
Disk, Memory, Network, Multimedia & Mobile forensics
Malware & Threat
Static/Dynamic analysis, YARA, Cuckoo, PE analysis
AI for Forensics
Deepfake detection, AI-assisted investigation, GNN anomaly detection
Tools & Frameworks Used
Industry-standard Deep Learning libraries and Digital Forensics toolkits.
65+ Latest MTech Computer Science Project Topics (2026)
Deep Learning + Digital / Cyber Forensics focused topics with recommended tools. All projects include IEEE base paper, source code, report, PPT and viva support.
| # | Project Title | Domain | Tools / Stack |
|---|---|---|---|
| 🧠 Deep Learning & Computer Vision | |||
| 1 | Vision Transformer (ViT) based Medical Image Classification and Segmentation | CV / DL | PyTorch · timm · MONAI · ViT |
| 2 | Real-time Object Detection and Tracking with YOLOv8 / YOLOv9 on Edge Devices | CV | Ultralytics YOLO · OpenCV · TensorRT |
| 3 | Self-Supervised Learning for Image Representation with SimCLR / MAE | Self-Supervised | PyTorch · MAE · SimCLR · Lightly |
| 4 | Multimodal Vision-Language Model for Image Captioning and Visual QA | Multimodal | PyTorch · CLIP · BLIP-2 · LLaVA |
| 5 | Efficient Transformers for High-Resolution Image Segmentation (SegFormer / Mask2Former) | Segmentation | PyTorch · MMSegmentation · Detectron2 |
| 6 | Few-Shot and Zero-Shot Learning for Rare Object Detection | Few-Shot | PyTorch · Detectron2 · CLIP |
| 7 | 3D Object Detection and Point Cloud Processing with PointNet++ / Point Transformer | 3D Vision | PyTorch · Open3D · PointNet++ |
| 8 | Action Recognition in Videos using Temporal Transformers / SlowFast Networks | Video | PyTorch · MMAction2 · SlowFast |
| 📝 NLP, Large Language Models & Multimodal | |||
| 9 | Domain-Adaptive Fine-Tuning of LLMs for Legal / Medical / Financial Text | LLM | Hugging Face · LoRA · PEFT · QLoRA |
| 10 | Retrieval-Augmented Generation (RAG) System with Vector Databases | RAG | LangChain · FAISS / Chroma · LlamaIndex |
| 11 | Aspect-Based Sentiment Analysis and Opinion Mining with Transformer Models | NLP | Hugging Face · BERT · RoBERTa · spaCy |
| 12 | Named Entity Recognition and Relation Extraction for Knowledge Graph Construction | IE | spaCy · Hugging Face · Neo4j |
| 13 | Multilingual Machine Translation with Efficient Transformer Architectures | MT | Fairseq · Hugging Face · OPUS |
| 14 | Instruction-Tuned Small Language Models for Edge Deployment | SLM | Hugging Face · LoRA · llama.cpp · GGUF |
| ✨ Generative AI & Diffusion Models | |||
| 15 | Conditional Image Generation and Editing with Stable Diffusion / ControlNet | Diffusion | Diffusers · ControlNet · Stable Diffusion |
| 16 | Deepfake Detection using Spatio-Temporal and Frequency Domain Features | Deepfake | PyTorch · Xception · EfficientNet · FaceForensics++ |
| 17 | Text-to-Image and Image-to-Image Synthesis with Latent Diffusion Models | GenAI | Diffusers · CompVis · Accelerate |
| 18 | Audio Deepfake Detection and Voice Spoofing Countermeasures | Audio Forensics | PyTorch · RawNet · AASIST · ASVspoof |
| 19 | Controllable Text Generation and Style Transfer with Large Language Models | Text Gen | Hugging Face · PEFT · Guidance |
| 🔍 Digital Forensics & Disk / File System Analysis | |||
| 20 | Automated Disk Image Analysis and Timeline Reconstruction with Autopsy / TSK | Disk Forensics | Autopsy · The Sleuth Kit · Plaso |
| 21 | File Carving and Recovered Artifact Classification using Machine Learning | File Carving | PhotoRec · Scalpel · Python · scikit-learn |
| 22 | NTFS / EXT4 Journal and Metadata Forensics for Deleted File Recovery | FS Forensics | TSK · Autopsy · Custom parsers |
| 23 | Cloud Storage Forensics: Artifacts from Google Drive, Dropbox and OneDrive | Cloud Forensics | Autopsy · Browser Artifacts · API logs |
| 24 | Mobile Device Forensics: Android / iOS Logical and Physical Acquisition Analysis | Mobile | Autopsy · ADB · libimobiledevice · Andriller |
| 💾 Memory Forensics | |||
| 25 | Windows / Linux Memory Dump Analysis with Volatility 3 for Malware Detection | Memory | Volatility 3 · Rekall · WinDbg |
| 26 | Process Hollowing and Code Injection Detection via Memory Forensics | Memory | Volatility · YARA · Custom plugins |
| 27 | Encrypted Memory Artifact Extraction and Ransomware Key Recovery | Memory | Volatility · Bulk Extractor · Strings |
| 28 | GPU Memory Forensics for Cryptocurrency Mining Malware Analysis | Advanced | Volatility · CUDA tools · Custom scripts |
| 🌐 Network Forensics & Traffic Analysis | |||
| 29 | Network Traffic Anomaly Detection using Graph Neural Networks | Network + AI | PyTorch Geometric · Wireshark · CICFlowMeter |
| 30 | PCAP Analysis and Protocol Anomaly Detection with Wireshark + Machine Learning | Network | Wireshark · tshark · Scikit-learn · Zeek |
| 31 | DNS Tunneling and Covert Channel Detection using Deep Learning | Network | PyTorch · Zeek · Wireshark · PCAPs |
| 32 | Encrypted Traffic Classification without Decryption using Flow Features + DL | Encrypted Traffic | PyTorch · CIC-IDS · NFStream |
| 33 | IoT Network Forensics and Device Fingerprinting | IoT Forensics | Wireshark · Zeek · ML classifiers |
| 🦠 Malware Analysis & Threat Intelligence | |||
| 34 | Static and Dynamic Malware Classification using CNNs and Transformers | Malware + DL | PyTorch · PEFile · Cuckoo · Ember |
| 35 | YARA Rule Generation and Automated Signature Creation for Malware Families | YARA | YARA · Cuckoo · VirusTotal API |
| 36 | Ransomware Behaviour Analysis and Decryption Key Recovery Techniques | Ransomware | Cuckoo · Volatility · IDA / Ghidra |
| 37 | Android Malware Detection using Static Permissions and Dynamic Behaviour | Mobile Malware | Androguard · CuckooDroid · PyTorch |
| 38 | Polymorphic and Metamorphic Malware Detection with Graph Embeddings | Advanced | PyTorch Geometric · Ghidra · NetworkX |
| 39 | Threat Intelligence Platform: IOC Extraction and Correlation from Reports | Threat Intel | MISP · OpenCTI · YARA · STIX/TAXII |
| 🖼️ Multimedia & Image/Video Forensics | |||
| 40 | Image Forgery and Splicing Detection using Deep Learning and Noise Patterns | Image Forensics | PyTorch · Noiseprint · ManTra-Net · OpenCV |
| 41 | Video Tampering and Frame Deletion / Insertion Detection | Video Forensics | PyTorch · OpenCV · FFmpeg · Optical Flow |
| 42 | Source Camera Identification using Sensor Pattern Noise (PRNU) | Camera ID | Python · OpenCV · PRNU extraction |
| 43 | Deepfake Video Detection with Spatio-Temporal and Physiological Signals | Deepfake | PyTorch · FaceForensics++ · MesoNet · Xception |
| 44 | Steganalysis: Detection of Hidden Data in Images and Audio using CNNs | Steganalysis | PyTorch · SteganoGAN · SRNet |
| 🤖 AI-Assisted Digital Forensics & Investigation | |||
| 45 | AI-Powered Timeline Reconstruction and Event Correlation for Investigations | AI Forensics | Plaso · Python · NLP · Graph analysis |
| 46 | Automated Artifact Triage and Relevance Ranking using Machine Learning | AI Forensics | Autopsy · scikit-learn · Custom ranking |
| 47 | Natural Language Query Interface for Forensic Databases (NL-to-SQL / Graph) | NL Interface | Hugging Face · LangChain · SQLite / Neo4j |
| 48 | Cross-Device Correlation of Artifacts using Graph Neural Networks | Graph + Forensics | PyTorch Geometric · Autopsy · Neo4j |
| 49 | Explainable AI for Malware Detection and Forensic Decision Support | XAI | SHAP · LIME · Captum · PyTorch |
| 🚀 Advanced & Research-Oriented Topics | |||
| 50 | Federated Learning for Privacy-Preserving Malware Detection across Organisations | Federated | Flower · PyTorch · Differential Privacy |
| 51 | Adversarial Attacks and Defences on Deep Learning based Forensic Classifiers | Adversarial | Foolbox · ART · PyTorch |
| 52 | Continual / Lifelong Learning for Evolving Malware Families | Continual | PyTorch · Avalanche · EWC / Replay |
| 53 | Graph-based Detection of Advanced Persistent Threats (APT) from Logs and Traffic | APT | PyTorch Geometric · Zeek · Elastic |
| 54 | Multimodal Fusion for Social Media Forensics (Text + Image + Metadata) | Social Media | PyTorch · CLIP · Transformers · EXIF |
| 55 | Zero-Shot and Open-Set Recognition for Unknown Malware / Attack Detection | Open-Set | PyTorch · OpenMax · CLIP |
| 56 | Efficient On-Device Deep Learning for Mobile Forensics Triage | Edge AI | TensorFlow Lite · ONNX · MobileNet |
| 57 | Blockchain and Immutable Logging for Forensic Chain-of-Custody | Blockchain | Hyperledger / Ethereum · Hashing · Smart Contracts |
| 58 | Privacy-Preserving Digital Forensics with Homomorphic Encryption / Secure MPC | Privacy | PySyft · TenSEAL · MP-SPDZ |
| 59 | Large-Scale Log Mining and Anomaly Detection for Insider Threat Detection | Insider Threat | PyTorch · Elasticsearch · Spark |
| 60 | Synthetic Dataset Generation for Training Robust Forensic Classifiers | Data Gen | Diffusion · GANs · StyleGAN · Augmix |
| 61 | Cross-Modal Retrieval for Linking Text Reports to Multimedia Evidence | Retrieval | CLIP · FAISS · Hugging Face |
| 62 | Automated Report Generation for Digital Forensic Investigations using LLMs | LLM + Forensics | LangChain · Llama / Mistral · Templates |
| 63 | Real-time Malware Detection on Network Flows using Lightweight Neural Networks | Real-time | PyTorch · NFStream · ONNX Runtime |
| 64 | Forensic Analysis of Container and Kubernetes Runtime Environments | Cloud Native | Falco · Volatility · kubectl · CRI-O tools |
| 65 | End-to-End AI-Assisted Digital Investigation Platform (Ingestion → Analysis → Report) | Platform | Autopsy · Volatility · PyTorch · LangChain · Dashboard |
★ All 65 MTech Computer Science project topics are sourced from IEEE Xplore, CVPR, ICCV, NeurIPS, ACL, USENIX Security, DFRWS, Digital Investigation journal and leading open-source forensic tool documentation (2022–2026). Each project includes the base paper, complete source code, datasets or generation scripts, architecture / investigation flow diagrams, university-format report for VTU / Anna University / JNTU, PPT (20–25 slides) and 50+ viva Q&A specific to the topic.
FAQ — MTech Computer Science Projects (DL + Forensics)
Computer Science Project Lab — Bangalore
Deep Learning workstations with GPUs, Digital Forensics lab (Autopsy, Volatility, Wireshark, Cuckoo), malware analysis sandbox, and dedicated mentoring rooms for MTech and PhD CSE scholars.
Workstations
& TSK
Memory
& Zeek
Sandbox
& Signatures
Detection
Pipelines
Preparation
Sessions